JPS6354675A - Method and device for generating picture data for continuous mode picture recording - Google Patents

Abstract

PURPOSE:To prevent the effect of a tone jump in a continuous picture record by changing tone edges like a pulse respectively in a multiple arrival element and/or its adjacent element and making the space of the pulse array change corresponding to the change of the tone edges. CONSTITUTION:An integer part IX steppedly changes from an initial value IY(=G0) and a decimal part DX steppedly and sequentially changes in pieces by a picture element and turns to 0 or its adjacent value in the position where the integer part IX changes. Moreover the total value FX of the decimal part DX finely and steppedly changes to enlarge parabolicly and every time it arrives 1. it returns to the value shown in FXi=FX(i-1)+DXi-1. In the position where the total value FX returns to the value of the formula a carry signal CRX occurs and the interval of its occurrence is sequentially shortened to the position where the integer part IX changes. After the integer part IX changes the carry signal CRX occurs at long interval again and the interval of the occurrence is sequentially shortened. The sparse period coincides with the changing period X of the integer part IX.

Description

Detailed Description of the Invention (Industrial Field of Application) This invention is aimed at recording an image whose gradation changes continuously using a pixel array having discrete gradation. Change (hereinafter referred to as [tone jump])
The present invention relates to a method and apparatus for generating image data that can prevent the influence of

(Prior Art and its Problems) In an image recording device such as a plate-making scanner, spatial gradation changes of an image must be expressed by a pixel array having discrete gradations. For this reason, Fig. 29 (a
), even if you want to record a continuous-tone image in which the gradation changes continuously according to the position change on the recording screen, step-like gradation as shown in FIG. It is necessary to approximately represent such a continuous tone image by an image with changes. This situation is common not only to plate-making scanners but also to devices that record images using digitized image data.

On the other hand, in currently used image recording apparatuses, various improvements have been made to make such discontinuous changes in gradation, that is, the tone jump ΔG shown in FIG. 29(b), more visible. Therefore, in general image recording, this tone jump ΔG is rarely visually recognized.However, in the following cases, the existence of this tone jump ΔG is relatively clearly visible.

The first case is a case where the gradation changes monotonically in accordance with a specified gradation change rate, such as when recording a slant screen in a plate-making scanner. In such cases, the smoothness of tone changes is often inhibited by the tone jump ΔG.

The second type is an image, such as an image of human skin, whose gradation changes gradually and which tends to attract the attention of the viewer of the recorded image. In this case as well, the presence of tone jump ΔG is easily recognized as disturbance in the recorded image.

The third case is when an original image is enlarged to create a large-screen recorded image. At this time, the size of the pixel increases due to the enlargement process, so if detailed observation is performed with the naked eye, the existence of tone jump ΔG will be detected as HH.

Various countermeasures have been considered to deal with such problems, and representative ones among them will be explained below along with their drawbacks.

(1) The data length of the image data, that is, the number of bits of the digital signal expressing the gradation, is made sufficiently large to reduce the absolute value of the tone jump ΔG.

This method is in principle the most desirable method.

However, in reality, the data m increases as the data length increases, and the processing time also becomes considerably long. Further, since it is necessary to prepare a circuit for processing such a large amount of data, the size and cost of the device will increase.

(2) The digitized image data is first converted into an analog signal, and natural noise such as circuit noise added to this analog signal is used. In this method, the analog signal with added noise is converted back into a digital signal, and this is used for image recording. Then, it is made impossible to specify the pixel where the tone jump ΔG occurs due to the above-mentioned natural noise or the like. However, since the level of natural noise is generally small, the number of bits in the digital image data must be quite large to apply this method. Therefore, it has the same drawbacks as the first method.

(3) Digital random noise having a certain level is generated by a noise generation circuit and added to digital image data. This method is different from the first and second methods described above, and can be expected to have some effect by simply adding a relatively simple circuit. However, in this method, noise unrelated to the properties of the image to be recorded is added to the image data, so there is a problem in that the image data cannot be corrected in accordance with the position where the tone jump ΔG occurs.

In addition, in Japanese Patent Application Laid-open No. 1 [61-91661, gradation signal (
By adding a fluctuation signal (for example, a sawtooth wave or a triangular wave) to a continuous tone signal, images are recorded in a discrete number of gradations. However, in this case, the peak value and wavelength of the fluctuation signal must be appropriately selected depending on the rate of increase or decrease of the gradation signal.

The reason for this is that the time when the gradation signal on which the fluctuation signal is superimposed exceeds the threshold value must match the pixel interval for recording as much as possible, so that effective recording can be achieved.

As described above, in the prior art, there is a problem in that it is difficult to effectively eliminate the influence of tone jump ΔG when recording an image in which the gradation changes continuously.

OBJECTS OF THE INVENTION The present invention is intended to overcome the above-mentioned problems in the prior art and provides an image that can effectively prevent the effects of tone jumps in continuous tone image recording without significantly complicating data processing. The purpose of the present invention is to provide a data generation method and device.

(Means for Achieving the Object) In order to achieve the above-mentioned object, the image data generation method according to the first invention of this application converts an image whose gradation changes continuously to a discrete gradation. Targeting an image data generation method for recording using a pixel array, first, the method includes an integer part having a data length corresponding to the number of 1llllrlll gradations used for image recording and a decimal part having an arbitrary data length. The first gradation data expresses the gradation of the image to be recorded for each pixel.

Then, the values of the decimal part are sequentially accumulated along the direction of the pixel array, and each pixel whose accumulation result reaches a value corresponding to a multiple of a predetermined reference value is defined as a multiple reaching pixel.

Further, each time the multiple reaching pixel exists, the value obtained from the integer part of the first gradation data for the multiple reaching pixel and/or its neighboring pixels is changed by a predetermined integer to obtain a second gradation. data is generated, and the second gradation data is used as image data for recording the image.

The image data thus obtained has, for example, spatial gradation changes as shown in FIG. 1(e), which will be described later. The reason why such image data can be obtained with the above configuration will be explained in detail in the description of the embodiments.

In addition, in a second invention of this application, as an apparatus for realizing the first invention, (1) an integer part having a data length corresponding to the number of gradations used for image recording and an arbitrary data length; a decimal part, and generates first gradation data chronologically in scanning order, in which the gradation of an image to be recorded is expressed for each pixel by the integer part and the decimal part; Means: ■ Sequentially accumulating the values of the decimal part along the direction of the pixel array, and generating a pulse of a predetermined size each time the accumulation result reaches a value corresponding to a multiple of a predetermined reference value. (2) second gradation data that generates second gradation data by chronologically synthesizing the pulse and the value obtained by 19 from the integer part of the first gradation data; and generating means, and uses the second W and tone data as image data for recording the image.

Note that "the cumulative result reaches a value that is a multiple of the reference value" refers to not only the case where the cumulative result matches these values, but also the case where the cumulative result jumps over these values. It is a concept that includes Moreover, it does not matter whether the multiple reaching pixel is specified directly or indirectly.

(Embodiments) A. First Embodiment FIG. 2 is an overall configuration diagram of a cylindrical scanning plate-making scanner incorporating an image data generation device according to a first embodiment of the present invention. The overall structure and general operation of this scanner will be explained below with reference to FIG. 2, and then the characteristic parts of this embodiment will be explained in detail.

(A-1) Overall configuration and general operation of the first embodiment - The plate-making scanner 1 shown in FIG. 5 is fixed to the drum shaft 6. This drum @6 is rotated synchronously by the driving force of a motor 9 applied through a pulley 7 and a belt 8.

On the other hand, a pickup head 10 and an exposure head 11 are respectively disposed so as to be movable parallel to the drum shaft 6, facing the original image drum 3 and the recording drum 5, and these are driven by a driving pulse motor 12. .13 to the feed screws 14 and 15 respectively,
They are being transferred to each other.

The pickup head 10 has a built-in photoelectric conversion element and the like, and reads pixels of the original image 2 to generate an image signal. This image signal is given to the A/D converter 16. In this A/D converter 16, this image signal is sampled in accordance with a periodic pixel clock CKX from a timing control circuit 24, which will be described later, and sequentially A/D converted. The image signal after A/D conversion is then given to the image processing circuit 17.

The image processing circuit 17 includes a gradation correction circuit (not shown) and performs predetermined processing such as gradation correction on the input image signal. The recording image signal obtained in this manner is applied to the halftone dot generator 18. Halftone dot occurrence 'a18
compares this recording image signal with the basic halftone dot signal to generate a halftone dot output, and the exposure head 11 records a halftone dot image on the photosensitive material 4 based on this halftone dot output. Of course, continuous tone recording may be performed based on the recording image signal.

In the case of halftone dot recording, one halftone dot is represented by, for example, 2.3 pixels×2.3 pixels, with one small square representing the area of one recording pixel in FIG. 16, which will be described later. Therefore, an image signal to be recorded is given to each pixel, and the image signal is recorded with a different dot percentage for each pixel. For this reason, 2
．． Halftone dots expressed by 3 pixels x 2, 3 pixels are circles, ellipses,
It is well known that not only regular shapes such as squares and diamonds, but also distorted shapes have been recorded.

On the other hand, this prepress scanner 1 is provided with a gradation gradient image generation circuit 20 as an embodiment of the present invention.

When recording a gradient mesh on the photosensitive material 4, the image data generated by the gradient gradient image generation circuit 20 is supplied to the halftone dot generator 18 via the image processing circuit 17. ing. At this time, the image signal read from the original image 3 is not used. However, depending on the application, these may be synthesized.

The prepress scanner 1 is also provided with a control circuit 21 connected to a keyboard 22 and an operation panel 23. This control circuit 21 is for controlling each of the above-mentioned circuits and performing various data processing. Furthermore, this scanner 1 is provided with a timing control circuit 24. This timing control circuit 24
The pixel clock CKX, the main scan start clock CKY that indicates the exposure start position for each rotation of the main scan, and , driving pulse motor 12.
13 (indicated by broken lines in the figure) is generated and applied to each part shown in the figure.

(^-2) l! i! ivA tilt image generation 20(7)
P+(7)n FIG. 3 is a block diagram showing the internal configuration of the gradation gradient image generation circuit 20. Moe) Just like that,
This gradation gradient image generation circuit 20 is a circuit that generates image data for recording a gradient mesh on the photosensitive material 4 shown in FIG. 2, and an example of the gradation distribution of this gradient mesh is schematically shown in FIG. is shown. That is, in the area 4a on the photosensitive material 4 in FIG. 4, an inclined mesh (hereinafter referred to as "main scanning direction inclined mesh") whose gradation changes monotonically along the main scanning direction X is shown. In addition, in the area 4b, an inclined mesh whose gradation changes monotonically along the sub-scanning direction Y (hereinafter referred to as "sub-scanning direction inclined mesh")
It is shown. Further, in the region 4C, a slope network whose gradation changes along an arbitrary direction P is shown.

As will be explained later, the gradation gradient image generation circuit 20 shown in FIG. 3 can freely generate image data for recording gradient meshes in various directions. ing. This image data is then generated in a manner that effectively removes the effects of tone jumps.

Therefore, in the following, we will first explain the operation of generating image data for recording a diagonal mesh in the main scanning direction using the circuit shown in FIG. Generation of image data for each of the gradient dot recordings will be explained. However, in the explanation of main scanning direction inclined mesh generation, general matters required for later explanation will also be described.

In the following explanation, 1' in decimal notation is used to distinguish ``1pa'', which is an integer value in decimal notation, from "1", which indicates the state of a specific bit of data.

is expressed as “1.”. Similarly, °2'' as an integer value of the 10 submodule is expressed as 1 lill.

To generate the main scanning direction inclined mesh, first, the operator uses the keyboard 22 shown in FIG. 2 to specify the main scanning direction X as the direction in which the gradation changes.

The following data are also input (see Figure 5).

m #f’l network generation same starting position X3 Inclined network generation end position X.

(2) Gradation initial field G at inclined network generation start position X.

(3)l! l! i: Parameter that indicates the J4 rate of change (the slope of the gradation change straight line in Figure 5) When these data are input, the control circuit 21 in Figure 2
A microcomputer (not shown) installed inside the
The amount of gradation change Qx per pixel in the main scanning direction X is calculated and determined. Hereinafter, this gradation change amount QX will be referred to as the "main scanning direction change amount."

The main scanning direction change mqX obtained in this way is
The signal is applied to a main scanning data generation circuit 100 provided in the gradation gradient image generation circuit 20 shown in the figure.

The main scanning data generation circuit 100 is a circuit that generates gradation data (first gradation data) for each pixel along the main scanning direction X, and performs an accumulation operation to be described later.

Also, the initial gradation value G input from the keyboard. is given to the sub-scanning data generation circuit 200. This sub-scanning data generation circuit 200 includes the main scanning data generation circuit 10 described above.
It has a configuration similar to 0. The sub-scanning data generation circuit 200 also receives an input of the amount of gradation change per pixel in the sub-scanning direction Y (hereinafter referred to as "the amount of change in the sub-scanning direction") gY. However, in the case of the generation operation of the main scanning direction inclined mesh considered here, Q7=0.

Among the various data thus given, the gradation initial value G. is taken into a sub-scanning direction adder 210 provided in the sub-scanning data generation circuit 200. Further, the sub-scanning direction change ff1qY (=O) is also latched by the latch circuit 220 and then taken into the sub-scanning direction adder 210. On the other hand, the main scanning direction change mqX is latched by the latch circuit 120 and then taken into the main scanning direction adder 110 provided in the main scanning data generation circuit 100.

(A-3b) The main scanning direction adder 110 is shown in Fig. 6.
2 is a block diagram showing the internal configuration of this main scanning direction adder 110. FIG. Of the circuits included in this main scanning direction addition Z110, the addition RK111 data selector 112 and latch circuit 113 are capable of processing data having a data length corresponding to the data length of image data used for image recording. There is. Therefore, for example, when gradation is expressed using M bits in image recording, these circuits are also formed as circuits that process M bits.

On the other hand, the adder 114 and data selector 11 in FIG.
5 and the latch circuit 116 are capable of processing any predetermined data length (for example, N bits). As will be explained later, the carry signal CR1 output from the adder 114 is transmitted to the adder 1.
As a result, the adder 110 in the main scanning direction can process (M+N) bit data formed from the upper M bits and the lower N bits. For example, M and N are 8
Each bit is allocated.

Among these, the upper M bits are used for gradation expression during image recording, and the least significant bit corresponds to the smallest unit of gradation expression. Therefore, the upper M bits have a meaning as an integer part of image data or gradation data. Furthermore, the lower N bits are not directly expressed in the image data finally obtained, and the gradation gradient image generation circuit 2
Used only within 0. Therefore, the lower N bits have a meaning as a decimal part of the gradation data.

(A-3C) The configuration of the sub-scanning direction adder 210 in FIG. 7 is as follows:
4 is a block diagram showing the internal configuration of the sub-scanning direction front zero device 210 of FIG. 3. FIG. As can be seen from a comparison between FIG. 6 and FIG. 7, this sub-scanning direction adder 210 has the same configuration as the main-scanning direction adder 110 of FIG. 6, and the difference between the two is the data handled. Only the contents and operation timing are shown.

Also, in the sub-scanning direction adder 210 in FIG.
An adder 211, a data selector 212, and a latch circuit 213 for processing an M-bit integer part, and an adder 214, a data selector 215, and a latch circuit 216 for processing an N-bit decimal part are provided. .

(A-36) Relationship between the integer part and the decimal part The relationship between the integer part and the decimal part is shown in FIG. That is, in FIG. 8(a), the integer part IX and the decimal part DX input to the main scanning direction adder 110 are M pit data: 8M-1'''°=al
-80 and N Got data: b,...b,b.

It is expressed by N-11. Moreover, in FIG. 8(b),
Integer part I in sub-scanning direction adder 210.

Similarly, the decimal part DY is M-bit data: CH-1
・°°°・01・00 and N-bit data: d, ... d, d.

Each is expressed by.

(A-38) Operation of the sub-scanning data generation circuit 200 Under such conditions, first, the sub-scanning data generation circuit 200
The operation will be explained. Sub-scanning direction adder 2 in Fig. 7
The initial gradation value G input in 10. is data selector 2
12 as its B input. In response to using an M-bit signal as image data for image recording,
This gradation initial value G. It also consists of M bits. Further, as a selection signal in the data selector 212, a recording start signal ST given from the control circuit 21 in FIG. 2 by pressing a start switch (not shown) provided on the operation panel 23 in FIG. 2 is used. Ru. and,
When the recording start signal ST is given, the data selector 21
2 is the B input, that is, the gradation initial value G. is selected and output to the latch circuit 213. This latch circuit 21
3 latches the input data and sends it to adder 211 at 7.
Give the input to . The B input of this adder 211 is set to 00...00'' (M bit ) is given.Furthermore, when the gradation is successively lowered along the sub-scanning direction Y, 11...11" (N bits) are given to the B input. Therefore, when a gradation change is applied only in the main scanning direction X as considered here, the B input of the adder 211 is OO...00''.

This adder 211 takes into account the carry signal CR2 from the adder 214 being used, adds the input to it and the 8 inputs, and provides the addition result to the input of the data selector 212. After the start of operation, the recording start signal ST maintains an inactive level, so that the data selector 212
selects the input to it and outputs it. This output is newly latched by the latch circuit 213.

In this way, the gradation initial value Go includes the B input of the adder 211 and the carry signal CR from the other adder 214.
2 are added repeatedly.

The repetition period is determined by the latch timing of the latch circuit 213, and the main scanning start clock CKY is used as the latch signal that determines this latch timing. Therefore, the above-mentioned repetitive operation is performed only once per main scan.

On the other hand, the amount of change in the sub-scanning direction q is input to the B input of the adder 214.
, is given. The addition RH 214 adds this B input to the input input from the latch circuit 216, and provides the addition result to the input of the data selector 215. However, if a carry occurs due to addition, the carry signal CR2 is applied to the adder 211, as described above.

The B input of the data selector 215 always has 00...
00" (N bits) is given. Then, when the recording start signal ST as a select signal is given, this B input is selected, and in the case of a square, the Δ input is selected. In this way, The selected data is latched by the latch circuit 216 and then given to the input of the adder 214.The cycle of such repetition is determined by the main scanning start clock CKY as the latch signal of the latch circuit 216. .

That is, at the start of operation, B of the data selector 215 is
The input is taken in and this repetition loop is initialized, and the N-bit change amount q in the sub-scanning direction is sequentially increased by +J[] tJ every time one main-scanning circuit is performed. Then, when a carry occurs in Kti[214,
The carry signal CR2 causes this carry to be carried to the integer part I,
reflected in Further, the output of the latch circuit 216 is output as a fractional part.

To summarize the above, it can be explained as follows.

In other words, in the sub-scanning direction addition 3210, first, as shown in FIG.
Take in 0 as the value of the integer part I. At this time, the decimal part is initialized to O11. In the next main scanning cycle, the O1 scanning change factor Q is added to the decimal part DY to become Dy-Qy (FIG. 9(b)). In the next main scanning cycle, the sub-scanning change mQ is further added to the fractional part, resulting in D=2QY (FIG. 9(C)). However, when 2g≧1, a carry occurs and the integer part I
, becomes (Go+1.), and the decimal part becomes (2CJY-
1,). Thereafter, such operations are repeated. However, in the case of the main scanning direction inclined mesh considered here, 9
．． = 0, the decimal part Dy remains "0" even as the 01 scan progresses, and the integer part I is G. It remains as it is.

FIG. 10A is a diagram showing the above-mentioned operation from another viewpoint in the case of q,≠O (an example will be described later). As can be seen from this figure, each pixel is defined by the main scanning coordinate X and the nI scanning coordinate y.

y), the decimal part for the first pixel P (1, 1), P (, 1, 2), ... on each main scanning line is displayed as follows for each scanning line. Scanning change ff
It changes by 1 qy. When the accumulated value reaches "1 degree", a carry signal CR2 is generated and the value of the decimal part DY returns to (accumulated value - 1,). However, in Figure 10A, the decimal part DY is "'0
” is shown.

On the other hand, the integer part I increments by (+1.) each time it receives the carry signal CR2.

Integer part I and decimal part obtained in this way.

The gradation data formed by this becomes the basis for generating "first gradation data" which will be described later.

Furthermore, as already explained, in the case of the generation operation of the inclined mesh in the main scanning direction, Qy = "0", so as shown in FIG. 10B, the decimal part is always "0"', and the integer part is I is always G. It is. Therefore, in the case of the main scanning direction inclined network, from the sub scanning direction adder 210 in FIG.
. ,D,="O" continues to be output.

The roof 260 formed by the data selector 2301 latch circuit 240J3 and adder 250 provided in the sub-scanning data generation circuit 200 in FIG. It has the ability to accumulate the fractional part outputted from the adder 210. When the cumulative result reaches a value that is 1 bit (2') more than the maximum value that can be accommodated by the N-bit adder 250, that is, ii 1,n, the carry signal CRY is output from the σ adder 250. It will be output as a pulse. When such a carry occurs, the output of the adder 250 becomes (accumulated value -1,), and then the accumulation starts again.

Therefore, in this roof 260, the cumulative result of the decimal part DY is II l, which is a multiple of 11 (including "1." itself).
It has the function of a circuit that outputs a pulse every time it reaches . The "pulse" output from this loop 1260 is the carry signal CRY.

However, in the case of generating an inclined mesh in the main scanning direction, D = "0", so no matter how many decimal parts are accumulated, the cumulative result will never reach "1. Therefore, in this case, the carry signal CRY is not generated.
A random number RD from a random number generation circuit 30 is given as an input.

This random number RD is used as the initial value for the above accumulation, but
The reason for providing such a random number RD will be described later.

(^-3f) Main/scanning-overtime rotation times 100 = On the other hand, the integer part I given from the sub-scanning direction adder 210 is the scanning direction adder n 110 in the main scanning direction data generation circuit 100. is given as an initial value for each scanning line. This main scanning direction adder 110 is connected to the sixth adder 110 as described above.
It has the configuration as shown in the figure. And that Noh is the 7th Noh.
It is almost the same as the aperture of the sub-scanning direction adder 210 shown in the figure, and differs only in data content and operation timing. Therefore, the main scanning direction adder 110 in FIG. 6 performs the following operation.

First, adder 114. The loop formed by the data selector 115 and the latch circuit 116 is initialized to "O" (N bits) by applying the main scanning start clock CKY as a select signal to the data selector 115. After that, this loop changes the main scanning direction change ff
i Q X is sequentially added to the initial value pixel by pixel. When a carry occurs in the N-bit adder 114, this adder 114 generates a carry signal CR1, and outputs this carry signal CR1 to the other adder 111.

On the other hand, adder 111. Another loop formed by the data selector 112 and the latch circuit 113 takes in the integer part I given from the sub-scan data generation circuit 200 as an initial value for one main scan.

Then, the integer part 8 is circulated in this loop in synchronization with the pixel clock CKX, and when the Yayari signal @CR1 is given, the integer part Ix is incremented by (→-1,). However, ALL is applied to the B input of addition E'3111.
"0'' is given.

Therefore, the integer part IX and the decimal part DX output from the main scanning direction adder 110 change along the main scanning direction X as shown in FIG. 11A. The gradation data for each pixel formed from the integer part IX and the decimal part Dx is "first gradation data" in this embodiment. As mentioned above, in the case of the operation of generating a screen tilted in the main scanning direction, even if the sub-scanning progresses, the 01 scanning direction Y
The integer part I, for , remains unchanged. For this reason, the 11th
The changes shown in Figure A are common to each main scanning line.

The decimal part Dx obtained in this way is added to the adder 1 in FIG.
given to 50. And data selector 130. A loop 160 formed by a latch circuit 140 and an adder 150 accumulates this fractional part DX in synchronization with the pixel clock CKX. The initial value for this accumulation is the N-bit random number RD given from the random number generator 30.
is used. The effect of using such a random number RD as the initial cumulative value will be described later, but for the time being,
The explanation will proceed assuming that the initial value of the accumulation is "0".

As shown in FIG. 11A, along the main scanning direction X, the decimal part Dx changes as 0, q, 2g, 3(lX,... If the values are accumulated sequentially along the scanning direction X, the accumulated value Fx initially increases sequentially, as shown in FIG. 12(a).

Then, when the cumulative value [FX reaches "1°", a carry occurs, and a carry signal CRX is sent from the adder 150 in FIG.
is output as a pulse, and the accumulated value Fx1 for the next pixel is FXi""FX(i-1)"DXi'・"°
It becomes (1). However, FX(i-1) and Dxi are defined as follows.

F −: Cumulative value X (i 1) just before carry occurs DXi ' Fractional part D for the pixel where carry occurs
Therefore, by repeating such an operation, a pulse train PT consisting of a chain of carry signals CRX as shown in FIG. 12(b) is generated.

As can be seen from FIG. 12(b), the pulse interval of the pulse train PT gradually decreases. This is because the value of the decimal part Dx to be accumulated increases sequentially, so that the accumulated value Fx becomes 1. This is because the interval until reaching ``'' becomes gradually narrower.

FIG. 13 illustrates this from a different perspective. Of these, FIG. 13(b) shows the accumulated value Ax obtained by sequentially accumulating the decimal part Dx of each pixel along the main scanning direction X. However, unlike the above-mentioned cumulative value FX, this cumulative value A is drawn in a rapid succession without worrying about carry. Moreover, in FIG. 13(a), the standard value is 11.degree. A numerical scale is drawn showing integers that are multiples of IP: ``i, 11.''12, H, , +i n, 11. Furthermore, cumulative glflA
Fractional parts corresponding to pixels where X reaches these integers are shaded.

On the other hand, as described above, the adder 150 of FIG. 3 sequentially accumulates the fractional part D and generates the carry signal CRX every time the cumulative value reaches "1." therefore,
The adder 150 indirectly specifies pixels having a diagonally shaded decimal part ("multiple reaching pixels") in FIG. 13(b), and outputs a carry signal CRX to those multiple specific pixels. It will have a function.

As described above, the pulse interval of the pulse train PT formed by the carry signal CRX becomes progressively narrower as the decimal part Dx increases, as shown in FIG. 13(C).

Then, as shown in FIG. 11A, when a carry occurs in the decimal part DX itself to be accumulated, the integer part IX becomes (+1
．． ), and the decimal part D is 111.11.
1ia (for example, "○"). Therefore, the change in density of the pulse train PT in FIG. 13(C) is repeated every time the integer part IX changes.

The above situation is shown in Figures 1 (a) to (d). However, in Figure 1, for convenience of illustration, small step-like changes in pixel units are omitted by broken lines. .

In FIG. 1, the integer part IX has an initial value of 1. (=G.

) to change in a stepwise manner (Fig. 1(a)). Also,
The decimal part DX gradually changes step by step in pixel units, and returns to 0'' or a value near it at the position where the integer part Ix changes (Fig. 1(b)).Furthermore, the decimal part D The cumulative value FX increases parabolically while changing in fine steps, and reaches "1. '', the value returns to the value shown by equation (1) (FIG. 1(C)). However, only a part of the waveform is depicted in FIG. 1(C).

Furthermore, in FIG. 1(C), the period until the accumulated value F8 returns to the value of equation (1) becomes progressively narrower toward the position where the value of the integer part IX changes.

Then, at the position where the accumulated value FX returns to the value of equation (1), the carry signal CRX shown in FIG. 1(d) is generated. The generation interval of this carry signal CRX is the integer part ■,
becomes gradually shorter toward the position where it changes. After the integer part (2) 8 changes, the carry signal CRX begins to be generated again at wide time intervals, and thereafter, the generation interval gradually becomes narrower. This density period corresponds to the change period ΔX of the integer part IX.

(A-3 (1) Combining operation of integer part and carry signal (pulse) In this way, the main scanning direction adder 110 in Fig. 3 outputs the integer part lx in Fig. 1(a), and adds vessel 150
The carry signal CRX shown in FIG. 1 ((1) is output from
is given to eight inputs. In addition, carry signal CRX +
, t The carry signal CR, ! passes through the OR circuit 42. :Become. In the case of the main scanning direction gradient network, the carry signal CRY which is the other input of the OR circuit 42 is always II O11, so the carry signal CR which is the output of the OR circuit 42 is
has the same data value as the carry signal CRX from adder 150. Then, this carry signal CR is A
It passes through the ND circuit 43 and becomes the least significant bit (LSB) input signal among the six inputs in the adder 41.

As the other input of the AND circuit 43, the adder 41
A signal obtained by inverting the carry signal CR3 from the inverter 44 is provided. Therefore, unless a carry occurs in the adder 41, the output of the inverter 44 is "1."
The output of the AND circuit 43 is the carry signal CR (=
CRX).

Therefore, the adder 41 outputs the integer part IX and the carry signal C.
It has a function of synthesizing R (=CRX,) to generate "second gradation data".

As a result, the image data S, which is the addition output of the adder 41, has a waveform as shown in FIG. 1(e).

As can be seen from FIG. 1(e), the image data S thus obtained is a step signal S corresponding to the integer part Ix. The pulse train PT is placed on top of the pulse train PT. The pulse density in this pulse train PT is
Step signal S. The level increases sequentially toward the position where the level increases by (, +1.). Therefore, when recording a main scanning direction inclined mesh using such image data S, macroscopically, the gradation appears to be changing continuously. Therefore, if such a method is used, the influence of tone jumps on recorded images can be effectively prevented.

Note that when a carry occurs in the adder 41 due to the addition of the carry signal CRX shown in FIG. 3, the output of the AND circuit 43 becomes o°. Therefore, at this time, the adder 41
The input to "OII" becomes "OII," and the value of the integer part IX becomes the output S of the adder 41. Therefore, in this case, image data S as shown in FIG. 14(a) is obtained. If no configuration is used, the configuration shown in Fig. 14(b)
As shown in , the effect of overflow may occur at the upper limit of gradation expression 1i112'.
It is desirable to provide an ND circuit 43 or the like.

(A-3h) Significance of random numbers as cumulative initial values and operation of random number generation circuit 30 Next, the random number R generated from the random number generation circuit 30 in FIG.
The reason for using D as the cumulative initial value will be explained.

As described above, by using the gradation gradient image generation circuit 20 of FIG. 3, the influence of tone jumps can be effectively prevented. However, the pulse train PT added to prevent the influence of tone jump has a common arrangement for each main scanning line.

Therefore, if schematic symbols such as those shown in FIG. 15 are used to indicate gradations, the image data processed by the above method will have a pixel arrangement as shown in FIG. 16A.

As can be seen from FIG. 16A, in this image, the pulses generated from the carry signal CRX are arranged in a straight line along the sub-scanning direction Y. Then, there remains a possibility that these straight lines can be recognized visually. In order to cope with this, in this embodiment, the initial cumulative value for accumulating the decimal part Dx is set to the random number RD for each main scanning line.

FIG. 17 is a block diagram showing the internal configuration of the random number generation circuit 30 of FIG. 3. In FIG. 17, when the main scanning start clock CKY is applied to the counter 31, the counter 31 increments by (+1) and outputs the count value to the ROM 32 as address data. A series of random numbers are stored in advance in the form of a table in the ROM 32. Then, one random number RD is output from the specified address.

This random aRD serves as 8 inputs to the data selector 130 shown in FIG. Then, when the main scanning start clock CKY is given as a select signal to the data selector 130, the input on this day is selected and given to the latch circuit 140. After that, during the period until the next main scan starts, the data selector 130
The Δ input given from 0 is selected and output. Therefore, the random number RD is used as a common value on one main scanning line, and a new random number is generated and used each time scanning moves to the next main scanning line.

In this way, by using the random number RD as the initial value for the accumulation operation of the fractional part DX, the carry signal CRX
The timing of occurrence of this changes for each scanning line. Therefore, as shown in FIG. 16B, from the carry signal CRX to t
The pulses q are dispersed in units of scanning lines, thereby making it possible to prevent the pulses from arranging in a straight line along the D1 scanning direction Y.

In the above, an inclined mesh in which the gradations increase along the main scanning direction That is, first, the day input of the adder 111 included in the front t9B110 in the main scanning direction in FIG. 6 is set to "11...11" (M bits). Then, this adder 111 receives a carry signal CR.
Every time 1 is given, the above data "" is stored in the integer part IX.
11...11'' is added.

As is well known, ``°11...11'' (M bit) is side-entered, and the 2's complement of this data, 00...01'' (M-bit), is subtracted. is equivalent. Therefore, when the carry signal CR1 is not generated, the integer part Ix is decremented by (-1,). Conversely, when the carry signal CR1 is generated, 11...11'' CR1 = ""11...1 1" + '"00...
・01′.

Since ="oo...00" holds true, the integer part Ix maintains the value just before that value.

On the other hand, in this case, as the amount of change qx in the main scanning direction, an absolute (two's complement) number to be decreased along the main scanning direction X is given. For example, if you want to decrease the gradation by "00...01'' (N bits) as a decimal value for each pixel, use (]x="11...11")
(N bits). Then, the adder 11 in FIG.
4 is substantially equal to °“00...01” (N
bits). As a result, the adder 114 repeats the ti operation as follows.

"00・=-00"+[;], = "11-11
”“11…11″10=”11…10”
+CR1× "11...10"+CI="11...00"+CR
100...01"+Q ="00...00''
+CR1”00・00” + to l, = “11-11”
Therefore, in this example, the carry signal C is used for every 2N pixels.
A pixel that does not generate R1 appears, and the integer part ■× in that pixel is decremented.

By changing the integer part ■X and the decimal part Dx in this way, the waveform diagram corresponding to FIG. 1 becomes as shown in FIG. 18. Therefore, even in the case of such a tone decreasing line,
The influence of tone jumps can be effectively prevented.

(A-4) Generation of sub-scanning direction inclined network The operation of generating the sub-scanning direction inclined mesh will now be explained.

When generating a sub-scanning direction inclined mesh, the main-scanning direction change ff1c+ (is set to "0"), and the sub-scanning direction change ff1c+ is set to a value (≠゛) according to the gradation change rate. Then, a loop consisting of an adder 214, a data selector 215, and a latch circuit 216 provided in the sub-scanning adder 210 in FIG. 7 synchronizes with the main-scanning start clock CKY. This sub-scanning direction gradient network Q is accumulated.

As a result, the decimal part DY changes as follows as the sub-scanning progresses: 0, Q, 2Qy, 3Qy, . . . Also, KafL? ! : When a carry occurs in 214, the output of the other adder 211 is (+1,)
or changes by (-1,), thereby integer part I,
also increment or decrement. Therefore, in the case of a slope mesh in which the gradation increases along the sub-scanning direction Y, the integer part 1 and the decimal part DY of the first pixel of each main scanning line change as shown in FIG. 10A described above. do. This fractional part is then sequentially applied to the IJD calculator 250 in FIG.

This adder 250 accumulates the decimal part via the data selector 230 and the latch circuit 240. However, the timing of this accumulation process is determined by the latch circuit 24.
It is determined by the pixel clock CKX given to 0. Therefore, the decimal part input to the R adder 250 is
While the value itself remains constant until scanning moves to the next main scanning line, the cumulative IUF of the decimal part D changes for each pixel. Therefore, the fractional part DY for each scan line
changes as shown in FIG. 19(a), and the cumulative value F for each pixel changes as shown in FIG. 19(b). However, in FIG. 19(b), RDo.

RDl, . . . are random numbers given from the random number generation circuit 30 in FIG. 3 for each scanning line.

On the other hand, the main scanning direction adder 110 in FIG.
The data is latched by the latch circuit 113 via the data selector 112 shown in the figure. However, since the main scanning direction change ff1c>x given to the adder 114 in FIG. 6 is 0'', the decimal part DX is always ''OII. For this reason, the adder 11
4, the carry signal CR1 is not generated, and as a result, the integer part I also maintains the value of 111]IY. Therefore, the decimal part Dx input to the adder 150 in FIG. 3 is always "O", and the integer part IX input to the adder 41 matches I. Furthermore, since Dx-' is "0", the cumulative value Fx of the accumulation loop 160 in FIG. 3 also maintains the value of the random number RD given as the initial value. As a result, carry signal CRX is not generated, and carry signal CR, which is the output of OR circuit 42, has the same waveform as carry signal CRY from nj scan data generation circuit 200.

The principle of generation of the carry signal CRY applied from the adder 250 to the OR circuit 42 in FIG. 3 has already been explained (A-3e
) as explained in the section. However, this adder 2
Although the accumulation loop 260 including 50 is included in the DJ scanning direction data generation circuit 200, it repeats the accumulation operation in accordance with the pixel clock CKX. Therefore, the carry signal CRY simply replaces the "main scanning direction
This results in a pulse train arranged according to a different rule from the pulse train obtained by reading the two. This will be explained in detail below.

First, on the first main scanning line, the first
As shown in FIG. 9(b), the cumulative value F is RDo (1,) regardless of the position X on the main scanning line. Therefore, the carry signal CRY is never generated. Next, on the second main scanning line, as also shown in FIG. 19(b), the cumulative value F increases by DY (-Q,) for each pixel, with RDl as the initial value. A carry signal CRY is generated every time this cumulative value F reaches II i , 11. This situation is shown in FIG. 20A. However, in FIG. 20A, the pixel in which the carry signal CRY is generated is indicated by an r x J mark. Also, for convenience of explanation, it is referred to as gy-2-4,
Random number RD is ignored.

Since D = 2gY on the third main scanning line (the straight line at the position Y=2 in Fig. 20A), the cumulative value FY is 2 x 2- for each pixel along the main scanning direction 4=2
' increase by '. Therefore, on this main scanning line, 1/2
= F = "1.7 for every 8 pixels, and the carry signal CRY is generated at each pixel as indicated by the "x" mark in Fig. 20A.
occurs. The same applies to the main scanning line of the pond.

As can be seen from FIG. 20A, the arrangement of the carry signals CRY along the sub-scanning direction/direction Y does not necessarily mean that the intervals become gradually narrower, but are arranged with partial spacing. There is. However, this is microscopic l15! In this case, when viewed macroscopically, the pulse intervals become progressively narrower.

Therefore, such a carry 5'3CRY is
When combined with the integer part I in the adder 41 shown in the figure, image data S shown in FIG. 21, for example, is obtained along the sub-scanning direction Y.

Therefore, even in the case of the mouth, the influence of tone jumps can be effectively removed.

As can be seen from FIG. 20A, in this case, the pulses of the carry signal CRY will not be aligned in a direction parallel to the main scanning direction X.

However, since the cumulative value F Y G is calculated according to a predetermined rule, as shown in FIG. There is a possibility of lining up. Therefore, in this practical example, the random number RD from the random number generation circuit 30 in FIG. Randomness is imparted to the distribution of the signal CRY.As a result, image data as shown in FIG. 208 can be divided by 1!7.

Note that if a modification such as giving the main scanning start clock CKY to the latch circuit 240 in FIG. 3 is applied, the same waveform as shown in FIG. 1 will be obtained along the sub-scanning direction Y. . Even if such a modification is made, no particular problem arises.

(A-4) Generation of a slope mesh in an arbitrary direction Next, generation of a slope mesh having a tone change direction in an arbitrary direction P (FIG. 4) instead of in a direction parallel to the scanning direction will be explained. In this case, a finite value (QXf-0, 0, ≠0>) corresponding to the gradation change direction is given to both the main scanning direction change amount q and the sub-scanning direction change fEj Q y. As the integer part I changes along Y, the integer part IX in the main scanning direction changes with this integer part ■ as the initial value for each main scanning line.
Two types of carry signals CRX and CRY are generated independently. A carry signal SCR containing both of these carry signals CRX and CRY is generated in the OR circuit 42 in FIG. 3, and this 11-tree signal CR is combined with the integer part Ix to form image data S.

Therefore, as shown in a schematic three-dimensional diagram in FIG. 22, the pulse group PG has a high density in the flat mesh portion FD below the step of the gradation jump point J and has a low density in the flat mesh portion FU above the step. Image data S having the following values is obtained. As a result, tone jumps are removed macroscopically, and a sloped network with smooth spatial gradation changes is provided.

FIG. 23 is a partial schematic diagram of a plate-making scanner incorporating an image data generation device according to a second embodiment of the present invention. The configuration of the parts not shown in FIG. 23 is the same as that of the scanner shown in FIG. 2, and a description thereof will be omitted.

In FIG. 23, this scanner has a configuration different from that of the first embodiment in the following points. First, the image data generation device according to the present invention is configured not as a gradation gradient image generation circuit but as a tone jump removal circuit 50 that uses image data obtained by reading from the original image 2 as input data. That's true. In other words, in the first embodiment, the image to be recorded (gradient mesh) was generated internally by the circuit, but in the second embodiment, when recording an image given from the outside, the tone We are trying to eliminate jumps.

Therefore, the tone jump removal circuit 50 in FIG. 23 is supplied with an image input that has been A/D converted by the A/D converter 60. Moreover, this tone jump removal circuit 50
The output is subjected to processing such as gradation conversion in the image processing circuit 17, and then output to the halftone dot generator 18. In addition, the 23rd
Although the gradation gradient image generation circuit 20 of FIG. 2 is not depicted in the figure, this is because it is unnecessary for the explanation of the second embodiment. Therefore, there is no problem in incorporating the gradation gradient image generation circuit 20 into the apparatus shown in FIG. 23 in the same manner as in FIG. 2.

A second feature of the apparatus shown in FIG. 23 is that a high-resolution A/D converter capable of converting (M+N) bits of digital data is used as the A/D converter 60. However, as in the first embodiment, M is the data length of the image data to be output, and N is the data length of the decimal part, which will be described later. This is because, in this second embodiment, the integer part and the decimal part are extracted from the input image data, as will be described later.

(B-2) Configuration and operation of tone jump removal circuit 50 FIG. 24 is an internal configuration diagram of the tone jump removal circuit 50. First, the configuration of this circuit will be explained. In the figure, of input image data converted into (M+N) bit digital data by an A/D converter 60, the upper M bits are applied as an integer part I to six inputs of an M-bit adder 51.

On the other hand, the lower N bits are applied to the B input - of the N-bit adder 54 as a fractional part.

The adder 54, together with the data selector 52 and the latch circuit 53, forms an accumulation loop 57 for sequentially accumulating the fractional part for each pixel. When the main scanning start clock CKY is input, the data selector 52 selects 'O'' (N bit), which is the B input, as the initial value, and selects the output of the adder 54 during other periods. The output is output to the latch circuit 53. Therefore, the accumulation loop 57 performs the same function as the accumulation loop 160 in FIG.
The circuit shown does not use random numbers. When using random numbers, the output of the random number generation circuit may be input to the B input of the data selector 52.

When a carry occurs in the adder 54 in FIG. 24, a carry signal CRs is generated.
3 is applied to the least significant bit of the B input of the other adder 51 via the AND circuit 56. In addition, the above AN
The D circuit 56 and the inverter 55 which inverts the carry signal CR4 of the adder 51 and supplies it to the AND circuit 56 are the first
This is provided to prevent the adder 51 from overflowing, as in the embodiment.

Next, the input image data R shown in FIG. 25(a) is
The operation of this tone jump removal circuit 50 will be explained by taking as an example a case applied to the D converter 60. However, horizontal @t in FIG. 25 indicates time or a scanning distance corresponding to the time.

First, the input image data R converted into (M+N) bit digital data by the A/D converter 60 is shown in FIG.
) is separated into an M-bit integer part I and an N-bit fractional part I. However, in this FIG. 25(b), the shaded part corresponds to the decimal part.

This fractional part is accumulated for each pixel in an accumulation loop 57 in FIG. And its cumulative value F is 1. ” is output, the carry signal CR3 is output, and the cumulative value F is (
Return to the value of F-1).

Therefore, in this accumulation loop 57, the accumulated value of the decimal part is actually an integer value (11i, a multiple of ll).
Carry signal CR as a pulse signal every time the
It has the function of a circuit that outputs 3.

This carry signal CR3 is combined with the integer part I in an adder 51 in FIG. As a result, the image data S changes as shown in FIG. 25(C). This second
As can be seen from FIG. 5, in a region 71 where the level of input image data R increases with respect to time t, the decimal part also increases sequentially. As a result, the interval between occurrences of the carry signal CR3 gradually becomes narrower. On the other hand, input image data R
The opposite phenomenon occurs in region 72 where the level of is decreasing. For this reason, among the points where a tone jump occurs (for example, points 73 and 74), a portion 75 on the lower gradation side
．． A high density pulse is applied to 76. Furthermore, near the maximum point 77 of the gradation, a high intensity pulse is given, and near the minimum point 78, the pulse density becomes low.

Therefore, macroscopically, the image expressed by the gradation changes in FIG. 25(C) is the input image data R in FIG. 25(a).
The image is visually recognized as having smooth tone changes close to .

Further, as shown in FIG. 26(a), when the level of the input image data R is an integer value, the decimal part is O''
Therefore, the carry signal CR3 is not generated, and image data S that is the same as the input image data R is output as shown in FIG. 26(b).

Furthermore, when input image data R that maintains a constant non-integer level is given as shown in FIG. 26(C), the decimal part also maintains a constant value, so as shown in FIG. 26(d). , a periodic carry signal CR3 is superimposed on the integer part I. Therefore, this device not only has the effect of removing tone jump tn, but also has the effect of faithfully expressing halftones in areas where the input image data R is constant. become.

(B-3) About the insertion position of the device In FIG. 23, the A/[) converter 60 and the tone jump removal circuit 50 are inserted between the pickup head 10 and the image processing circuit 17. Processing circuit 1
7 and the halftone dot generator 1°8. In this case, the A/D converter 16 shown in FIG. 2 is not required. Naturally, the image processing circuit 17 uses an analog processing circuit.

C9 Modified Example Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and for example, the following modifications are also possible.

(2) When capturing one pulse train by accumulating the decimal parts, it is not necessary to use the carry signal as it is. For example, a circuit that generates a pulse having a desired size by inputting a carry signal may be provided, and the pulses thus obtained may be combined into an integer part. Therefore, it is possible to generate a pulse not at the multiple reaching pixel itself but at a pixel in the vicinity thereof. Further, a pulse may be generated in a small area (for example, an area of several pixels) including both of these. An example of this case is shown in FIG.

Furthermore, the pulse height is also 1. ′, for example, °゛2
．． ” is not prohibited.

Therefore, in the present invention, it is sufficient to change the value of the integer part by a predetermined amount m in a pulsed manner for the multiple reaching pixel and/or pixels in its vicinity.

■ To change the value of the integer part in a pulse-like manner,
As shown in FIG. 28, it may be changed in the direction of Ω. In this case, the two's complement of the decimal part is calculated and accumulated, thereby reversing the direction of density change of the negative pulse.

■ When specifying a pixel that has reached a multiple by accumulating the decimal part, the desired reference value and the accumulation (direct) may be compared using a comparator. It is not essential to set ILII as the reference value; Other values may be set. For example, if a circuit that can handle numbers in the range l 2 and 11 is used as the accumulation loop, and when the accumulation value reaches 12, it is generated. If a carry signal is used, 112 IT is set as the reference value.The pulse density in this case is half that of the above embodiment.

■ In the above example, the integer digits of the accumulated value are not counted;
An adder 150°250.54 capable of accommodating only decimal digits is used. By using IM+ as the reference value, without counting the integer digits of the accumulated value,
Carry signal CRX.

This is because pixels that reach multiples can be specified using only CRY and CR3. However, this does not prohibit counting the cumulative value to integer digits. It is also possible to use a subtracter to perform the accumulation and a borrow signal to generate a pulse.

Furthermore, as in the above embodiment, it is not essential to generate the image data S online, but it is also possible to obtain the image data S offline and store it in a large-scale memory such as a magnetic disk. . Transfer the image data S 1q in this way to another location via a communication line,
It is also possible to perform recording afterwards.

■ The embodiment shown in FIG. 2 uses the halftone dot generator 18, but
It is not essential to use the halftone dot generator 18, and continuous tone recording based on image data may be used.

(2) This invention is applicable not only to cylindrical scanning plate-making scanners, but also to plane scanning plate-making scanners, as well as facsimile machines and copying machines that have gradation reproducibility.

(Effects of the Invention) As explained above, according to the present invention, the gradation changes in a pulsed manner in each of the multiple reaching pixels and/or the pixels in the vicinity thereof, and the intervals of the pulse array correspond to the gradation changes. Therefore, it is possible to obtain an image data generation method and apparatus that can effectively prevent the effects of tone jumps in continuous tone image recording.

In addition, the image data generation process uses two types of data: an integer part and a decimal part, but the generated image data itself does not require a particularly large data length, so data processing becomes too complicated. Not at all.

[Brief explanation of the drawing]

FIG. 1 is a waveform diagram showing the process of obtaining image data according to the first embodiment of the present invention, FIG. 2 is an overall configuration diagram of a cylindrical scanning plate-making scanner incorporating the first embodiment, and FIG. The figure shows a gradation gradient image generation circuit 2 as a first embodiment.
0 is a block diagram that does not show the internal structure of 0, FIG. 4 is an explanatory diagram of the inclined mesh, FIG. 5 is an explanatory diagram of data input from the outside in J3 for generating the main scanning inclined mesh, and FIGS. 6 and 7. are block diagrams showing the internal configurations of the main scanning direction mouthpiece and the subscanning direction side mouthpiece, respectively; FIG. 8 is an explanatory diagram of the integer part and decimal part of the first gradation data; FIGS. 9 and 10A. , FIG. 10B, FIG. 11A, and FIG. 11B are explanatory diagrams of addition processing and carry of the decimal part. FIGS. 12 and 13 are explanatory diagrams of accumulation of the decimal part and carry signal generation. The figure is an explanatory diagram of overflow prohibition processing, Fig. 15,
16A and 168 are explanatory diagrams of the effect of random number provision, FIG. 17 is an internal configuration diagram of the random number generation circuit, and FIG. 18 is the operation of the first embodiment in generating a gradient network in which the gradations decrease sequentially. 19 to 21 are explanatory diagrams of image data generation operation for a sub-scanning direction inclined mesh, FIG. 22 is an explanatory diagram of image data for an inclined mesh having an arbitrary tone gradient direction, FIG. 23 is a partial configuration diagram of a plate-making scanner incorporating the second embodiment, FIG. 24 is a block diagram showing the internal configuration of the tone jump removal circuit 5o formed as the second embodiment, and FIG. 25 26 is a waveform diagram showing the operation of the second embodiment, FIGS. 27 and 28 are detailed explanatory diagrams of the present invention, and FIG. 29 is an explanatory diagram of tone jump. DESCRIPTION OF SYMBOLS 1... Cylindrical scanning plate-making scanner, 2... Original image, 20... Gradation gradient image generation circuit, 30... Random number generator, 50... Tone jump removal circuit, 100... Main Scan data generation circuit, 200...DI scan data generation circuit, 160.260
...accumulation loop, 1, I, I...integer part, × Y DX, D,...decimal part,

Claims (8)

[Claims] (1) A method for generating image data for recording an image whose gradation changes continuously by a pixel array having discrete gradations, the method comprising: generating image data according to the number of discrete gradations used for recording; The gradation of the image is expressed for each pixel by first gradation data including an integer part having a data length and a decimal part having an arbitrary data length, and the value of the decimal part is expressed in the direction of the pixel array. Each pixel whose cumulative result reaches a value corresponding to a multiple of a predetermined reference value is defined as a multiple reaching pixel, and for each existence of the multiple reaching pixel, the multiple reaching pixel and/or A value obtained from the integer part of the first gradation data for the neighboring pixel is changed by a predetermined integer to generate second gradation data, and the second gradation data is
An image data generation method for continuous tone image recording, characterized in that the image is used as image data for recording. (2) When an image in which the gradation changes continuously is an image in which the gradation changes in accordance with the specified gradation change rate along the specified direction of gradation change, the For the gradation data No. 1, the gradation value specified for a predetermined pixel is set as an initial value, and the amount of change determined according to the gradation change direction and the gradation change rate is added to the initial value in scanning order. An image data generation method for continuous tone image recording according to claim 1, wherein the image data is obtained by sequential addition or subtraction. (3) When an image whose gradation changes continuously is an image obtained by reading an original image, the integer part and the decimal part of the first gradation data refer to the gradation of the image of the original image. Generation of image data for continuous tone image recording according to claim 1, which is provided by upper data and lower data, respectively, when digitized with a greater number of gradations than the number of gradations in scanning recording. Method. (4) The image data is image data for scanning recording, and when accumulating the decimal part of the first gradation data, a random number is given as an initial accumulation value for each scanning line. An image data generation method for continuous tone image recording according to any one of items 1 to 3. (5) A device that generates image data for recording an image whose gradation changes continuously using a pixel array having discrete gradations, the device comprising: first gradation data including an integer part having a corresponding data length and a decimal part having an arbitrary data length, and expressing the gradation of an image to be recorded pixel by pixel by the integer part and the decimal part; , first gradation data generating means that generates chronologically in scanning order; and accumulating the values of the decimal part sequentially along the direction of the pixel array, and making the accumulation result a multiple of a predetermined reference value. an accumulating means for generating a pulse of a predetermined size each time a corresponding value is reached; and a second gradation data generating means for generating gradation data of 2, wherein the second gradation data is image data for recording the image. Image data generation device for. (6) When an image whose gradation changes continuously is an image whose gradation changes at a specified gradation change rate along a specified gradation change direction, the first gradation data generation The means sets a gradation value designated for a predetermined pixel as an initial value, and sequentially adds or subtracts a change 1 determined according to the gradation change direction and the gradation change rate to the initial value in scanning order. 6. The image data generation device for continuous tone image recording according to claim 5, wherein said image data generation device is a means for generating said first tone data. (7) When an image whose gradation changes continuously is an image obtained by reading an original image, the integer part and decimal part of the first gradation data refer to the gradation of the image of the original image. Generation of image data for continuous-tone image recording according to claim 5, which is provided by upper-side data and lower-side data, respectively, when digitized with a greater number of gradations than the number of gradations in scanning recording. Device. (8) The image data generation device further includes random number generation means, and when accumulating the decimal part of the first gradation data in the accumulation means, the random number from the random number generation means is used at the initial stage of accumulation for each scanning line. An image data generation device for continuous tone image recording according to any one of claims 5 to 7, which is given as a value.