JPH03219780A - Picture processing system - Google Patents

Abstract

PURPOSE:To keep an edge of a picture and to obtain the picture with less unsharp or the like by outputting any of a multi-value data latched by 1st and 2nd latch means in response to the state of an inputted noticed binarized picture element data so as to form a binarizing picture. CONSTITUTION:A selector 105 selects either of registers 103 (storing density 50) and 104 (storing density 200) depending on a level ('0' or '1') of a signal line 102 being an output signal from an input section 101 and outputs the result to a binarizing section 106 as an 8-bit signal. Naturally, when the level of the signal line 102 is '0' (=OFF picture element), the register 103 is selected and when the level of the signal 102 is '1' (=ON picture element), the register 104 is selected. That is, when a picture element of density '0' is received as a binary photoelectric conversion section data from an input section 101, a picture element of density '50' is outputted and when a picture element of density '255' is received as a binary photoelectric conversion section data from the input section 101, a picture element of density '200 is outputted. A binarizing section 106 applies binarizing processing using the density conservation type error spread method to a binary picture (density is '50' or '200') subjected to density conversion processing.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image processing method that receives a binary image and performs gradation processing.

[Prior Art] Conventionally, it has been rarely considered to perform image processing on a multi-valued image after it has been converted into a binary image. There is a method of performing binarization processing after gradation conversion to obtain a binary image.

Here, the general method for converting a binary image into a multivalued image is as follows:
It is as follows.

A window of a certain size is provided and a binary image is scanned.

Then, the number of dots placed within the window is counted, and the area ratio of the total area of the window to the number of dots is determined. Depending on the area ratio, the pixel of interest is converted into a pixel with multi-value data.

For example, if the window size is 4 x 4 (=16) pixels, there are 8 dots in the window, and the multilevel density is represented by 8 bits (256 gradations), then It is assumed that the pixel has a density of 256×(8/16)=128.

In the manner described above, the original multivalued image can be estimated from the binary image. Next, if you want to uniformly increase the density of the entire image (for example, if you want to uniformly subtract a certain value from the multilevel density obtained earlier and perform binarization processing such as error diffusion again,
As a result, a binary image with high density is obtained.
Here, it is defined that the lower the value of the multivalued data, the higher (deeper) the density.

[Problems to be Solved by the Invention] However, in the above-mentioned conventional technology, the average density is calculated for each region, and the same method as an image obtained by passing through a smoothing filter is obtained. Therefore, a problem arises in that the edge portions of the image or the entire image become blurred. In addition, there is also the drawback that the hardware scale becomes large due to multi-value processing.

The present invention has been made in view of such prior art, and it is an object of the present invention to perform gradation conversion of a binary image with a simple configuration, and to form a binary image based on the gradation-converted data. The aim is to provide an image processing method that makes it possible.

[Means for Solving the Problem] The image processing method of the present invention that solves this problem includes, for example, the following configuration. That is, an input means for inputting binary pixel data, and a first holding means for storing and holding multi-value data M corresponding to the maximum density value of the binary pixel data and at least less than the data indicating the maximum density as multi-value data. and a second holding means for storing and holding multi-valued data N corresponding to the lightest density value of the binary pixel data and at least equal to or greater than the data indicating the lightest density as multi-valued data; According to the state of the target pixel data, the first. and output means for outputting the multivalued data held in one of the second holding means.

[Operation] In the configuration according to the present invention, either of the multivalued data held by the first or second holding means is outputted by the outputting means depending on the state of the binary pixel data of interest inputted by the inputting means. Then, a binary image is formed based on the multivalued data outputted by this output means.

[Embodiments] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram of the image processing apparatus of this embodiment.

When binary image data (data obtained by converting multivalued data into binary data) is imported into this apparatus from the outside, it is first stored in the RAM within the input unit 101 . Thereafter, each pixel data of the binary image stored in the input section 101 is sequentially read out in the main scanning direction and output onto the signal line 102 (1 bit). Note that although this is a binary image stored in the input unit 101, the possible values are either "0" or "1".Since "0" pixels are pixels that are not printed during actual printing, they will be explained below. “OFF pixel”
, conversely, a pixel that is "1" is called an "ON pixel". Also, when the number of gradations reproduced by the image processing in the example is "256 (=88 bits)", the "OFF pixel" has a density of "0" and is ON.
"Pixel" has a binary density of "255".

The register 103 contains the density value when adding the value (C,) that makes the pixel with density "0" (=OFF pixel), and the register 104 contains the density value of the pixel with density "255" (=ON pixel).
The density value obtained by subtracting the value consisting of is stored in advance. In the embodiment, "5o" is stored in the register 103, and "5o" is stored in the register 103.
Assume that "200" is stored in 04.

The selector 105 selects one of the registers 103 and 104 according to the level (“0” or “l”) of the signal line 102, which is the output signal from the input section 101, and outputs it to the binarization section 106 as an 8-bit signal. do. Of course, at this time, when the level of the signal property 102 is "0 (=OFF pixel)", the register 103 is selected, and the level is "'l (=ON pixel)".
Then register 104 is selected. That is, when a pixel with a density of "O" is received from the input unit 101 as binary pixel data, a pixel with a density of "50" is output, and a pixel with a density of "255" is output.
”, pixel data with a density of “200” is output.

The binarization unit 106 performs a binarization process on the binary image (density value "50" or "200") that has been subjected to the density conversion process. The binarization process in the embodiment uses a dense preservation type error diffusion method.

A brief explanation of this error diffusion method is as follows.

The error generated when the binarization process is performed on the pixel consisting of the following pixels is distributed (distributed) by giving a predetermined weight to the positions of the unbinarized pixel group located in the vicinity of the binarized pixel position.

Here, errors that have been distributed in the binarization process up to that point are accumulated in each non-binarized pixel group. Now, when a pixel of interest is binarized, it is binarized based on the density value of the pixel of interest input from the outside, plus the error value accumulated up to that point at the pixel of interest position. It is something.

Note that, instead of using the error diffusion method, a dither method or a density pattern method may be employed. In any case, the present invention is not limited to the binarization processing by the binarization unit 106.

Now, the binarization unit 106 outputs a signal indicating the lightest density (=OFF pixel) or maximum density (=ON pixel) that the output unit 106 can reproduce.

As described above, according to the embodiment, the input density is 0.255.
The binary image is subjected to density conversion processing to convert it into a binary image with a density of 50.200, and this converted binary image is regarded as a multivalued image and re-binarized using the error diffusion method to have a density of 0.200. 2
By rearranging the dots into a binary image with 55,
You can change the gradation of the image. Moreover, unlike conventional methods, no smoothing processing is performed, so problems such as blurring of the edges of the image or the entire image do not occur.

Furthermore, there is also the advantage that this effect can be achieved with an extremely simple configuration.

Note that, in principle, processing may be performed as shown in the flowchart shown in FIG.

To explain briefly, first, in step Sl, “O”
Input a binary pixel of either “1” or “0”
”. After that, proceed to step S2, and convert the pixel “0” to “50” and the pixel “255” to “200”.
Convert to concentration value. That is, it is converted into a binary pixel that is either "50" or "200". Next, proceeding to step S3, the binary pixel data that can only take either "50" or 200" is treated as pixel data that can take values "0 to 255", that is, multivalued pixel data, using error diffusion method. Then, since the binarized data is obtained in step S4, a visible image or the like is formed based on it.

<Description of Second Embodiment> FIG. 3 illustrates a second embodiment based on the principle of the present invention.

In the second embodiment, consider performing gradation conversion on a color image composed of binary images of three colors: R (red), G (green), and B (blue).

In the illustration, binary data of R, G, and B are input from input signal lines 301, 302, and 303, respectively.
, G, and B and the selectors 310 to 312
．． Since the binarization units 313 to 315 are the same as in the first embodiment, their explanation will be omitted.

By setting various density values in registers 304 to 309, the contrast of the entire image can be changed,
You can change the tone of a certain color or freely adjust the tone of an image. For example, as shown in Figure 4, the register 30
Let us consider the case where a value is set to 4 to 309.

i,) When pattern P1 is selected, the contrast of the entire color image decreases.

if, ) In the case of pattern P2, the entire image has a yellowish tinge.

If, ) pattern P3, the entire image becomes clear.

You can control the image in this way.

Furthermore, this idea may be taken a step further by allowing the operator to change the value stored in each register to a value desired by the operator (for example, an "adjustment knob" may be provided outside the device).

In this example, image data of R2O and B was used based on additive color mixture, but C (cyan) was used based on subtractive color mixture.
Of course, data of 9M (magenta) and Y (yellow) may be used.

Description of Third Embodiment> FIG. 5 shows a block configuration of an image processing apparatus in a third embodiment.

Binary image data (data obtained when multivalued data is binarized) is input from an input line 501 in rask order and stored in a line buffer 502.

As shown in FIG. 6, the line buffer 502 is a FIFO that stores and holds binary pixel data for four lines in the main scanning direction.
It is composed of meso 9601 to 604, and is sequentially shifted in synchronization with a pixel clock (not shown). For example, FIFO
When focusing on the binarized data output from the memory 602, the binarized data input from the input line 101 is the data after two lines, and the data after one line is output from the FIFO memory 601. That will happen. Also, FI
From the FO memory 603, data from l lines before is transferred to the FI
The FO memory 604 will output data for two lines before. In other words, from the line buffer 102, binarized data for five pixels at the same position in the main scanning direction and continuous in the sub-scanning direction is sent to the peripheral determination circuit 502.
is output to. The peripheral determination circuit 503 performs a process of counting and outputting the number of dots placed in a 5×5 matrix window centered on the pixel position of interest, that is, the number of "ON pixels".

Since the maximum number of "ON pixels" is 25, a 5-bit signal is output to the gradation conversion lookup table 504. At the same time, the level of the pixel of interest (“0
' or "l") signal is outputted via a signal line 508 to an adder circuit 506, which will be described later.

Now, in the gradation conversion lookup table 504, according to the data output from the peripheral determination circuit 503, data (8 bits) is output by referring to the contents of the table. The register 505 also indicates that the pixel of interest (center of 5×5) is “O”.
The data to be added in the case of "N pixels" (in this case, 255
) are stored. Of course, it is possible to convert the data using a control means such as a CPU.

Addition circuit 506 basically adds the value output from gradation conversion lookup table 504 and the value output from register 505. The level of the signal line 508 is "1" when the pixel of interest is an "ON pixel" and is "0" when the pixel is an "OFF pixel." Adder circuit 506 switches its output according to the level of signal line 508. That is, when the signal line 508 is "L", the result of addition is output, and when it is "0", the result of addition is output ("
The binary conversion circuit 507 receives the value output from the addition circuit 506, converts it into a binary signal using an error diffusion method, etc., and outputs it to the output signal line 509.

Next, the peripheral determination circuit 503 will be explained using FIG.

The output from the line buffer 502 is added to the adders 701 to 70.
5 is input. Each adder has latches 707 and 708,
709, 710 and an adder 706, the pixel is shifted by one pixel each time the pixel transfer lock CLK 711 rises, and the adder 706 adds the sum of five consecutive pixels in the horizontal line. In this way, each addition section 701 to 70
5, the number of "ON pixels" in each consecutive five pixel group is counted. These are then taken into an adder 712, and the number of dots placed within a 5×5 pixel area centered on the pixel of interest is calculated. In the figure, the binarized data of the pixel of interest is stored in the latch 70 in the adder 703.
8, the signal is output as the signal line 508.

Next, the gradation conversion lookup table 504 will be explained in detail.

The data on the number of "ON pixels" (0 to 25) within the 5×5 window output from the peripheral determination circuit 503 is subjected to data conversion using a conversion table as shown in FIG.
It is input to an adder circuit 506. For example, if we consider a case where "5" is input as the number of "ON pixels", the conversion table outputs data "15". This lookup table is composed of RAM or ROM and can be rewritten, and by rewriting the contents, various gradation conversions can be performed.

Furthermore, as for the binarization circuit 507, a binarization method that preserves density, such as an error diffusion method, may be used, and the description thereof will be omitted in the present invention.

Based on the above processing outline, we will explain how data actually flows.

In FIG. 9, it is assumed that a binary image such as 901 is scanned by raster scanning, and each pixel is input from the input signal line 501 in synchronization with the pixel clock. In addition, in the illustration, the shaded pixels are “O”.
"N pixels", and the other white parts indicate "OFF pixels".

Now, the pixel of interest is number 90 of the binary image 901.
Assume that the position number is 3. That is, consider a 5×5 window 905 in which the pixel of interest is an “ON pixel” and is centered thereon. At this time, the window 905
The number of “ON pixels” included in the peripheral judgment circuit 503
, and “5” is output to the adder circuit 506.

Furthermore, it is converted into data "15" by the gradation conversion lookup table 504 based on the conversion table shown in FIG.

In the adder circuit 506, the data "255" set in the register 505 and the previous data "15" are added, resulting in data "270".

At this time, since the pixel of interest is an "ON pixel" as explained earlier, the value "270" is the value of the binarization circuit 507.
It is output to .

Furthermore, when the pixel of interest is 904, the 5×5 window 906 includes 6 “ON pixels”, so the gradation conversion lookup table 504 shows “2
5" is output, but the pixel of interest 904 is "
Since the pixel is an OFF pixel, the signal line 508 is at the "0" level, and data "0" is output to the binarization circuit 507.

As described above, when the data output from the adder circuit 506 is binarized by the binarization circuit, a gradation-converted image is obtained.

Description of the fourth embodiment> A fourth embodiment will be described using FIG. 10.

Binary image data is input from the input line 501 in rask order, enters the line buffer 502, and is divided into five groups including the pixel of interest.
The horizontal line is input to the peripheral determination circuit 503, and the number of "ON pixels" within the 5×5 window and the pixel signal of interest are output to the signal line 508, as in the third embodiment.

Further, the gradation conversion lookup table 1003 outputs data by referring to the contents of the table in accordance with the data output from the peripheral determination circuit 503.

Further, at this time, the table can be switched by setting the contents of the register 1001 from a control circuit such as a CPU.

Furthermore, in the third embodiment, the pixel of interest is "OF
As in the case of pixel ",", "0" is output, but in the embodiment of Sui-Tei 4, the gradation can be controlled by setting a value in the register 1002.

The selector 1004 switches between the output from the register 1002 and the output from the 1-gradation conversion lookup table as follows, depending on the level of the signal line 508.

That is, when the pixel of interest is an "OFF pixel" (signal line 508 is "O"), the output from the register 1002 is selected, and when it is an "ON pixel", the output from the gradation conversion lookup table 1003 is selected. Select output. Furthermore,
The binarization circuit 507 converts the value output from the selector 1004 into a binarization method that allows density preservation, such as an error diffusion method.
It is converted into a binary signal and output to the output signal line 509.

The line buffer 502 and the peripheral determination circuit 503 are the same as those in the third embodiment, so a detailed description thereof will not be given.

Gradation conversion lookup table 1003 and register 10
01 will be explained.

Two pieces of data are input to the gradation conversion lookup table 1003. One is O~ from the peripheral determination circuit 503.
This is the number data (5 bits) of 25 “ON pixels”,
The other is 3-bit data from register 1001. Depending on the data from the register 1001, it is possible to select which table among the tables shown in FIG. 11 is used for tone conversion.

The memory addresses of the gradation conversion lookup table 10o3 are mapped as shown in FIG. table T
3. Table T4. Table T5 is selected.

To explain with an example, when performing gradation conversion based on table T4, it is sufficient to set 3° in register 1001. At this time, there are 6 “ON” in the window.
If "pixel" is included, the gradation conversion lookup table 1003 points to the address "66H (H is a hexadecimal number)", and as a result, data "280" is output and gradation conversion is performed. Ru.

As described above, when configured as in the fourth embodiment,
You can easily switch between many types of gradation conversion.

<Description of the fifth embodiment> A fifth embodiment will be described based on FIG. 13.

In the fifth embodiment, it is considered that tone conversion is performed on a color image composed of binary images of three colors: R (red), G (green), and B (blue).

Binary data of R, G, and B comes in from input signal lines 1301, 1302, and 1303, respectively. R signal processing section 1307. G signal processing section 130B, B signal processing section 130
The configuration and data flow of 9 are the same as those of the third embodiment, so their description will be omitted.

Each R, G, B signal processing unit 1307, 1308° 1309
register 505, gradation conversion lookup table 5
By rewriting the contents of 04 in various ways, it is possible to change the contrast of the entire image, change the color tone, and freely adjust the tone of the image.

In the fifth embodiment, based on additive color mixture, R, G
, B was used, but based on subtractive color mixture, C(
Data for cyan), 9M (magenta), and Y (yellow) may also be used.

In addition, in the fifth embodiment, three processing sections, the R signal processing section, the G signal processing section, and the B signal processing section, were provided to flow data in parallel. Tone conversion of a color image can also be performed by sending R, G, and B data to the division.

[Effects of the Invention] As explained above, according to the present invention, a binary image is gradation-converted with a simple configuration, and based on the gradation-converted data (even if the binary image is formed, the edges of the image are not is preserved, and it is possible to obtain a good image with less blur.

[Brief explanation of drawings]

FIG. 1 is a block configuration diagram of an image processing device in the first embodiment, FIG. 2 is a flowchart showing the processing procedure in the first embodiment, and FIG. 3 is a block diagram of the image processing device in the second embodiment. 4 is a block diagram showing the pattern of numerical values set in the register in the second embodiment. FIG. 5 is a block diagram of the image processing device in the third embodiment. Figure 7 is a diagram showing details of the line buffer, Figure 7 is a diagram showing details of the peripheral determination circuit in Figure 5, Figure 8 is a diagram showing conversion characteristics of the gradation conversion lookup table in Figure 5, and Figure 9 is a diagram showing details of the peripheral determination circuit in Figure 5. Figure 10 is a block configuration diagram of the image processing device in the fourth embodiment; Figure 11 is the conversion characteristic of the gradation conversion lookup table in Figure 10. FIG. 12 is a diagram showing the storage state of table data for each conversion characteristic of the gradation conversion lookup table in FIG. 1O. FIG. 13 is a block configuration diagram of an image processing device in the fifth embodiment. It is. In the figure, 101...input section, 102...signal line, 10
3.104,304-309,505.1001 and 1
002...Register, 105, 310, 311 and 3
12...Selector, 106 and 313-315...
Binarization section, 107... Output section, 301 to 303...
Data input lines, 316-318...Data output lines, 5
01...Input line, 502...Line buffer, 50
3... Peripheral determination circuit, 504.1003... Gradation conversion lookup table, 506... Addition circuit, 50
7... Binarization circuit, 1307... R signal processing section, 1
308...G signal processing section, 1309...B signal processing section. Funeral map ^→- Funeral map Funeral map No. 7 Figure 01 Funeral map No. 1

Claims (3)

[Claims] (1) An input means for inputting binary pixel data, and a first storage for storing and holding multivalued data M corresponding to the maximum density value of the binary pixel data and at least less than the data indicating the maximum density as multivalued data. means, a second holding means for storing and holding multivalued data N corresponding to the lightest density value of the binary pixel data and at least more than the data indicating the lightest density as multivalued data; an output means for outputting the multivalued data held in one of the first and second holding means according to the state of the pixel data of interest. (2) An input means for inputting binary pixel data having either of the two states B0 or B1, and when the binary pixel data of interest inputted by the input means is in the B1 state, the binary value near the pixel of interest a correction means for correcting the value of the pixel data of interest by converting it into multi-value data according to the binary state of the pixel data group; and when the binary pixel data of interest inputted by the input means is in the B0 state, the pixel of interest is a conversion means for converting into multi-value data corresponding to the state of the data; and a conversion means for converting the multi-value data into multi-value data corresponding to the state of the data; An image processing method characterized by comprising an output means for outputting one of value data. (3) The image processing method according to claim 2, wherein the correction means includes a plurality of correction tables and a designation means for specifying a desired correction table among the individual correction tables.