JPH01286674A - Data synthesizing method - Google Patents

Abstract

PURPOSE:To prevent mixed data and its presence from appearing on a reproduced picture and to prevent a picture quality from being deteriorated by determining a dither matrix based on multilevel data obtained by quantizing the density information of an original picture and the desired data to be mixed. CONSTITUTION:Dither matrix is determined based on the multilevel data obtained by quantizing the density information of an original picture and desired data to be mixed and a desired dither picture and document data are outputted. Namely, a transmitter 1 is composed of a picture fetching part 3, a quantizing part 4, a density gradation part 5, a matrix selecting part 6, a document data fetching part 7, and a cell storing part 8 and an original picture 20 is inputted to a fetching part 3 and document data 21 is inputted to the fetching part 7. Here, a prescribed processing is executed, outputted from the selecting part 6, reproduced by a receiver 1, outputted as document data and the data picture is outputted as it is. For this reason, a receiver 2 is composed of a matrix detecting part 9, a cell storing part 10 and a character reproducing part 11.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a data synthesis method used when mixing or separating other data such as a document into an image signal.

(Conventional technology) In recent years, with the development of communication technology, OA (office
Networking of automation (automation) equipment is progressing, and the information handled is also progressing from mainly text data to including audio and image data.

However, among such data, document data,
Image data is generally transmitted and stored separately because the signal processing formats are different.

Furthermore, in transmitting information between networked OA devices, a confidential means is essential to prevent leakage to third parties.

In view of these circumstances, various methods have been proposed to mix information such as document data into image information.

One such example is Meiki and Yuki's ``Image Deep Encryption Using Arithmetic Codes'' (1986 Cryptography ε
1 Information Security Symposium Materials).

This converts the image data to 0.1, 2. ..., (arranged on n- one-dimensional coordinate axes, extract m coordinates (l≦m≦n) that correspond one-to-one to the data to be mixed from among these, and After calculating the exclusive OR of the image data located and the data to be mixed, and recording the result at each coordinate, the coordinate elements 0, 1,
2. ....

The (n-1) sequence is transmitted as an image signal.

If such a method is used, it is possible to handle both at the same time, which is not only extremely convenient, but also makes it possible to transmit even more important information in a confidential manner by making it appear as if the image is being sent electronically.

According to this method, important information can be kept secret, so it can also be used as a type of encrypted communication means.

However, with this method, the required image data directly changes depending on the other data to be mixed in, so the data that was mixed in in the relevant part on the playback screen appears as is, and it is obvious at a glance whether data has been mixed in or not. However, there was a drawback that the image quality was significantly degraded.

To compensate for this, the above literature proposes a method of convolving other data into the boundary area where black and white change on the image, but even if such a method is used, noise will be mixed on the screen. However, deterioration in image quality is unavoidable, and in order to reduce this, it is necessary to drastically reduce the amount of data mixed in, or to limit the amount of data mixed in only to the edges of the screen. It was impractical.

In view of these drawbacks, the present applicant has proposed a method of mixing information such as document data into image information using a dither method (Japanese Patent Application No. 62-47310).

In this proposal, we focus on the fact that the dither matrix (hereinafter referred to as a cell) can be configured in any way,
By setting cells corresponding to outside pixels based on the data to be mixed, desired data can be mixed into the image signal without impairing the pseudo gradation properties of the dithered image.

However, since this proposal mixes document data and the like into the binarized image, it has the disadvantage that it is not possible to increase the amount of document data that can be mixed.

(Object of the Invention) In view of the above circumstances, the present invention allows a large amount of data to be mixed into image data, and also prevents the mixed data and its existence from appearing on the playback screen. The object of the present invention is to provide a data synthesis method that can prevent image quality from deteriorating due to

(Summary of the Invention) In order to achieve this object, the present invention determines a dither matrix based on multivalued data obtained by quantizing density information of an original image and desired data to be mixed. It is characterized by mixing a large amount of desired data into an image signal without impairing the pseudo gradation properties of a multivalued dither image.

(Example) Hereinafter, the present invention will be explained in detail based on the illustrated example, but before that, in order to facilitate understanding of the present invention, a brief explanation will be given of the dither method, particularly the degree of freedom of cell setting in this dither method. Explain.

The dither method is a method that skillfully utilizes the integral effect, which is one of the properties of the human eye, to express intermediate densities using devices such as facsimiles and printers. The threshold value for determination is varied according to a predetermined rule, and the number of dots corresponding to the local average value of the original image III light is generated to express the medium rH1 density.

In this case, the method in which the threshold value is determined based on a random function or a pseudo-random function is called a random dither method, and the entire original image consisting of a matrix-like pixel array is further divided into subcells, and these subcells are The method of associating a predetermined dither pattern with a threshold value for each pixel is called the systematic dither method, but the latter is superior to the former in terms of resolution, gradation reproducibility, and amount of noise. .

When a grayscale image is quantized into L values using the m* dither method, and the cell size is nXn pixels, the number of pseudo gradations Leq that can be expressed is Leq = (L-1)Xn2
+1 ...... expressed as (1).

If the resolution of the display device is high, the visual integration effect is used, and the cell size is selected to the extent possible, the image can be expressed only with pseudo gradation for each cell, and the pixel arrangement does not depend on the original image. , can be determined arbitrarily. That is, it is considered that the method for determining the pixel arrangement has a degree of freedom depending on the size of the cell. Therefore, by determining the pseudo Pa key number Leq assigned to the cell from the density information of the original image, and further determining its pixel arrangement based on the character information, image information and character information can be combined into one data. be able to.

If this composite data is directly displayed or printed, it can be used as image data.

In addition, the pixel values in each cell that make up this composite data,
By focusing on the arrangement, it is possible to decode and use the synthesized character information.

-For example, if the quantization number of the image is set to "3" and the cell size n is set to "2", the pseudo-Fll key Leq that can be expressed by one cell is "9". , it is possible to express densities of "0" to "8" levels by one cell. In this case, as shown in FIG. 2, there is one cell 17 each showing density levels "0" and '8', but there are a plurality of remaining cells 18 showing density levels "1" to "7". ,exist.

That is, if the density level when one pixel is quantized is, for example, "3", 16 (24) combinations are possible, as is clear from FIG. 1O.

Therefore, when 4-bit character information is extracted from one document data, and this value is, for example, "12" in decimal notation, and the density level when one pixel is quantized is, for example, "3", If you select the 12th pixel array pattern that expresses density level "3" and send it,
On the receiving side receiving this signal, a quantized image based on a ternary density pattern can be obtained, and by paying attention to the pixel arrangement order of this received signal, character information "12" can be decoded. In other words, 4-bit character information can be mixed into this pixel.

The data synthesis method of the present invention based on the above-described image synthesis principle will be explained below.

FIG. 1 is a block diagram showing an example of a data transmission/reception system to which an embodiment of the data synthesis method according to the present invention is applied.

The data transmitting/receiving system L shown in this figure, the data transmitting device L
and a receiving device 2. When an original image 20 and document data 21 are manually generated, the original image 20 and document data 21 are synthesized by the transmitting device 1, and this synthesis result (transmitted data ) is transmitted to the chi receiving device 2 side.

Then, the receiving device 2 reproduces dithered image data and character data from the transmitted data, and supplies these to a next-stage device (not shown).

The transmitting device 1 includes an image capture section 3, a quantization section 4, a density N:A section 5, a matrix selection section 6, a document data capture section 7, and a cell storage section 8. 20
and document data 21 are input, these original images 20 and document data 21 are combined to create dithered image data, and this is transmitted to the receiving device 2 side.

The image capturing section 3 is equipped with an imaging device such as a CCD or a line sensor, reads the original image 20 when it is inserted, and supplies the reading result (image signal) to the quantizing section 4 .

The quantization unit 4 includes a threshold generator that generates a plurality of threshold values, a plurality of threshold values obtained by the threshold generator, and the image signal supplied from the image capture unit 3. It quantizes the image signal and supplies the quantization result (binary data) to the density gradation unit 5.

The density gradation section 5 is equipped with a PJ:Al11 generator, etc., and when the "binary" data is supplied, this "
Gradation rank data j is generated from the binary data and is supplied to the matrix selection section 6.

Further, when the number of extraction bits (data) b is supplied from the matrix selection section 6, the document data importing section 7 imports data from the document data 21 by the number of bits indicated by the number of extracted bits. Then, this is supplied to the matrix selection section 6.

In addition, the cell storage section 8 has an R cell storing section 8 in which a plurality of cells 17 and 18 corresponding to each concentration level m as shown in FIG. 2 are stored.
OM lt is provided, and when a read signal is supplied from the matrix selection section 6, the cell information stored at the address specified by the read signal is read out and supplied to the matrix selection section 6.

The matrix selection block 7 selects the density FIi harmonic J
When J is supplied, the density level m corresponding to this density gradation number J is calculated, and the number of extraction bits S corresponding to this density level m is calculated.

Then, this extracted bit number S is supplied to the document data importing section 7, and the data obtained at this time is inputted into the IO progress (
Convert to a numerical series Ki). After this, the concentration level m
and the cell storage unit 8 based on the numerical series K i
The cell 17 (or cell 18) is read out and transmitted to the receiving device 2 side as a dithered image.

The receiving device 2 includes a matrix detection section 9 and a cell storage section 1.
0 and a character reproducing unit 11, the transmitting device I
When a dithered image is supplied from , it is supplied as image data to a next-stage device (not shown), and the dithered image is decoded to reproduce character data, which is then supplied to the next-stage device. .

The cell storage unit IO includes a ROM storing the same cell information as the cell storage unit 8, and the matrix detection unit 9
When a read signal is supplied from the read signal, the cell information at the address indicated by the read signal is read out and supplied to the matrix detection section 9.

When the dithered image is supplied, the matrix detection section 9 detects the density level m of this dithered image, and while looking up the cell storage section IO, detects the cell 17 (or The cells 18) are sequentially read out and compared. When these patterns match, the cell number (binary data) is supplied to the character reproduction section 11.

When the character reproduction section 11 is supplied with the "binary" data, it stores it in sequence, and converts this storage result into 1
Character data is generated by dividing it into bytes (or l words), and is supplied to the next stage device.

Next, the encoding/decoding procedure of this embodiment will be explained with reference to the flowcharts shown in FIGS. 3(A) and 3(B).

(Encoding procedure) First, when the original image 20 is input, the image capture unit 3 captures it, and the quantization unit 4 converts this imaging result into 9
It is quantized into steps and supplied to the density gradation section 5.

As a result, the density gradation section 5 starts operating, generates Pa tone order data j from the "2'' base data outputted from the quantization section 4, and supplies it to the matrix selection section 6 ( Step 5TI).

The matrix selection section 6 converts the gradation order data j into a density level m while looking up the cell storage section 8, and checks how many cells there are corresponding to this density level m (step 5T2).

Here, if the density level m of this pixel is "3",
The matrix selection unit 6 selects the number of cells (
Extracted bit number data d, indicating "16" (value indicates 4 bits) is generated and supplied to the document data importing section 7.

The document data importing unit 7 extracts the extracted bit number data d,
Data is cut out from the beginning of the document data 21 by the number of bits indicated by and is supplied to the matrix selection section 6 (
Step 5T3).

The matrix selection unit 9 selects the document data 2 l tfi
At the same time, the data supplied from the decimal data d is converted into decimal data d (step 5T4), and a cell with a number corresponding to the decimal data d is selected from among the cells 18 corresponding to the density level m, and this is dithered. Receiving device a as image
2 tl! 4 (step 5T5).

In this case, if the value of the data cut out from the document data 21 is "1100" in "binary" notation, the document data importing unit 7 receives -12 in "IO" notation corresponding to the value ""1100".・is output.

Therefore, at this time, the "12" cell among the cells 18 corresponding to the density level m is selected and transmitted to the receiving device 2 as a dithered image.

Thereafter, steps 5TI-9T5 described above are repeated, and lii!
A dither image is generated in units of ii elements, and these are sequentially
It is transmitted to the receiving device 2 side.

Then, when all the original images 20 have been transmitted, this encoding process ends (step 5T6) (decoding procedure) Also, if the dithered image obtained by the above-mentioned operation is received by the receiving urIi2, this It is directly supplied to the next stage device as a dithered image.

Further, in parallel with this operation, the matrix detection section 9 detects the density level m of the dithered image and generates extraction bit number data d1 corresponding to this density level m.

Thereafter, the matrix detection section 9 accesses the cell storage section 10 based on the density level m, reads out the cell information corresponding to this density level m, and also reads the pattern of this cell and the pattern of the dithered image. Check if they match.

When these match, the cell number is converted into a binary number and supplied to the character reproduction section 11 (step ST11).

The character reproduction unit 11 receives “2” from the matrix detection unit 9.
``When base data is supplied, it is stored sequentially, and the stored data is 1 byte (or 1 byte).
Each time it becomes a word), it is converted into character data and supplied to the next stage device (step 5T12).

Thereafter, repeating step 5TIO-ST12 described above, each time a dithered image is received, it is directly supplied to the next-stage device, character data is reproduced from this dithered image, and this is transferred to the next device. Supply to the next stage equipment.

Then, when the transmission from the transmitter l is completed, this decoding process is completed (step 5T13).

In this way, in this embodiment, we focused on the degree of freedom in cell selection in the dither method, and selected cells corresponding to character data, so that the pseudo-N tonality of the multilevel dither image is not impaired. A large amount of character data can be mixed into the image signal without any problem.

In this case, since 5.3 gradation cells are used, more character data can be mixed in than when 2 gradation cells are used.

Regarding this point, referring to Figures 4 to 6 below,
It will be explained in further detail.

Now, the size of one cell 18 is nXn, and the concentration level displayed by this cell 18 is rn (rn
=Ot 1 m 2 y .
If the number 8 is expressed as C(m), then for this cell 18! 1! Once the numbers/values n and m are determined, the maximum value of the integer value d that can be synthesized in this cell 18 is determined.

Once the maximum value of this integer value d is determined, the number of bits that can express this integer value d, that is, the number of bits extracted from the document data, is determined.

In this case, bit nb, is b, = [Log2(C(m))] (bit
)−(2). Here, [X] represents the maximum ff number that does not exceed X.

In addition, the number of strokes of the density level m included in the l@image is W
When expressed as (m), the total number of bits that can be synthesized into this image B
is expressed as . Here, the number of combinations C(m) is determined by the number of physical gradations that can be displayed by the device and the size n of the cell, and the total number is the number of permutations that take n2 duplicates from L different ones. Since it is a total number, it holds true.

Further, assuming that all combinations of pixel arrays are used, the number of bits can be determined from equation (2) using the value of equation (4). As is clear from this equation (4),
It can be seen that as the number of physical gradations increases, the amount of data that can be synthesized increases exponentially.

However, in reality, when calculating the number of bits from the number of combinations C(m), truncation is performed for each gradation according to equation (2), so it is not possible to use all the combinations shown in equation (4). Can not.

Therefore, in reality, the amount of data that can be synthesized does not increase exponentially, so a method that does not include truncation errors is used to find the value of the amount of data that can actually be synthesized.

One method is to apply the number of combinations C(
m) and the number of bits, S, are determined, and then the average value of each density level, S, is determined.

For example, for the sake of simplicity, it is assumed that a display device capable of displaying three values of "0", "", and "2" and whose cell size n is "211" is used.

At this time, the number of pseudo gradations Leq that can be expressed is 9 levels (2・22+1) from equation (1), and the density levels m that can be pseudo expressed are '0', 'l', '2', ...
, "8".

Each of these density levels m is expressed by a combination of pixels shown in FIG. 4, and each density level m' at this time
The number of N combinations is C(m).

In this case, assuming that the probability of appearance of each value indicating the concentration level m is equal, the average IIfI of C(m) glass in FIG.
When finding b, D===n, 44 ・°
・(S)ノLeq 9 Similarly, the number of physical gradations becomes ・L=2・3,
When the size n is n = 2 + a, 4, the amount of information that can be synthesized in each cell and the amount of information contained in one pixel obtained by encoding are shown in Figure 5. The table shown below is obtained.

As is clear from this table, if the value of the cell size n is the same, the larger the value of the number of gradations, the larger the number of bits that can be synthesized.

In addition, in this embodiment, the original image 2 is
0 and the document data 21 to generate a dithered image, which is sent to the receiving device 2tl! Since the data is sent to II, the amount of data can be reduced compared to sending these individually.

This point will be briefly explained below with reference to FIG.

First, if the human output bit rate on the side of the transmitting device 1 used in this embodiment is determined, a table as shown in FIG. 6 can be obtained.

As is clear from this table, the number of quantization bits of a manually generated grayscale image is determined by the pseudo gradation number Leq of the above equation (1) for each pixel.
can be obtained as LOgz (L eq) by expressing (see Figure 6 (a) #I), and the amount of information that can be synthesized is as shown in Figure 5 (see Figure 6 (c) column) .

In addition, both the output code amount by the density pattern method and the output code amount by this method are n2・Log(
L eQ) bit. Here, assuming that the human power of the transmitting device 1 according to this embodiment is a quantized grayscale image and character data such as document data, the value of (output code amount)/(human power code ff1) is the 6th The values are shown in column (e) of the figure.

However, the dithered image output from the transmitter H position 1 is a 6th image that is a quantized grayscale image and displayed using the density pattern method.
If it is considered that the amount of information in FIG. 6(C) is synthesized without changing the code amount of W in FIG. 6(b), the above value becomes the value shown in the <r> column in FIG. 6. In other words, according to this embodiment,
Compared to the case where both data are input and output independently, the amount of code can be suppressed to about 84%.

FIG. 7 is a block diagram showing an example of the case where the embodiment of the data synthesis method according to the present invention is applied to encrypted communication means. In this figure, the same reference numerals are given to the parts corresponding to those in FIG. 1.

The system in this figure is different from the system shown in FIG.
6 is provided, and the document data 21 is encrypted and decrypted using keys 21 and 22.

In this case, the scrambler 15 changes the order of data read from the document data 21 based on the data supplied from 8!21.

Further, the descrambler 16 changes the order of data output from the character reproduction section 11 based on the data supplied from the key 22, and generates character data.

By doing so, the document data 21 can be transmitted or received in a confidential manner.

Furthermore, in each of the embodiments described above, the cell storage section 8.10
Although the cells representing each density level m are stored in the memory cell 1, the desired cells may be sequentially determined by calculation.

Various methods can be considered for this method. The method shown in the flowchart of FIG. 8 is one of them.

Hereinafter, a method for successively determining a desired cell will be explained based on this flowchart. Here, in order to simplify the explanation, it is assumed that the pseudo N function is "4311." and that the cell size n is "2".

First, the arithmetic unit @ (not shown) performs step 5T2.
0, the pixel of interest in the original image 20 is quantized and the density level m of this pixel is determined. After that, the process proceeds to step ST21, where the number of bits is calculated from this density level m based on equation (2) above. , is calculated. in this case,
If the density level m of the pixel is '3', the bit number S is '4', and as shown in FIG. 9, the combination of cells tkli4tc (m
) will be "16" pieces.

Next, in step 5T22, the arithmetic unit reads only Rb from the beginning of the bit string constituting the document data.
After cutting out the data, the process proceeds to step 5T23, where it is converted into a decimal number d.

After this, in step 5T24, the arithmetic device calculates the concentration "
The number of cells consisting of a combination of pixels showing 1", CI (,
CI=4) and the value of the 10 path number d, and this 1
It is determined whether the decimal number d can be expressed by a cell consisting of pixels with density "1".

Here, if the value of the decimal number d is d=o, 1,2°3, then d<aCt, so a cell consisting of pixels of (1,1,1°0), (1,l, A cell consisting of pixels of (0.1), a cell consisting of pixels of (1,0,1,1), (0,1,
1, 1), the arithmetic unit determines that the decimal number d can be expressed by any of these cells, and proceeds from step 5T24 to step 5T25.

Then, in step 5T25, the arithmetic device executes the arithmetic expression "d/4Ct" to obtain the quotient α and the remainder β.

In this case, the value of the lO base number d is d = 0.1, 2°3, and the value of the number of combinations aCs is "4 It, so
α=0, β:0 (or β:1.2゜3)
become.

Then, in step 5726, the arithmetic unit calculates this remainder β
According to the value, the pixel arrangement position of density "0" is determined, and the cell is determined.

In this case, if β: 0, out of each cell shown in FIG.
The first cell from the left (left in the figure) is calculated, and if β:1, the second cell from the leftmost is calculated. Similarly, if β: 2, the third cell from the left will be calculated, and if β: 3, the fourth cell from the left will be calculated.

Further, in step 5T24, if d≧4CI, the arithmetic operation a position is replaced with the arithmetic expression “d
−s=d a C+”.

After this, the arithmetic unit calculates the pixel indicating "Kandosan 2" and the density "
Number of cells consisting of a square combination with pixels showing 1" Act・3
C1 (AC+・3CL=12) is compared with the value of the lO base number d, and a cell is formed of a combination of a pixel where the lO base number dl indicates a density of ゛2°° and a pixel where the density indicates "l". Determine whether it can be expressed by

Here, since the density level m of the pixel of interest in the original image 20 is "3pa," the value of the lO base number d is "0" to "11".
”, and dl<, C, ·, C1, so the calculation H position is ``1'' pixel indicating density ``2'' and density ``
It is determined that this decimal number d can be expressed by any of the cells consisting of the pixel "1" indicating "1", and this step 5
The process advances from T24 to step 5T25.

Then, in step 5T25, the arithmetic device executes the arithmetic expression "d+/3ct" and obtains the quotient α and the remainder β.

In this case, the value of the decimal number d1 is any one of '0' to '11', and the value of the combination number 3CI is '3'', so the quotient α is any one of '0' to '3'. , and the remainder β is one of "0" to "2".

Next, in step 5T26, calculate! The position a is this quotient α
In accordance with the value of , a pixel arrangement position of density "2" is determined, and in accordance with the value of remainder β, a pixel arrangement position of density "1" is determined, and a cell is determined.

In this case, the quotient α and the remainder β are, for example, α=2,
If β=3, the 12th cell from the left (from the left in the figure) among the cells shown in FIG. 9 is calculated.

Further, in the decoding process, the inverse operation of the above-described procedure is performed from the pixel density of the cell determined in this way and its arrangement position, and the lO-adic number d synthesized in this cell is extracted.

If the cells are determined sequentially in this way, there is no need to store the cells, and the memory capacity can be significantly reduced.

Furthermore, in each of the embodiments described above, each bit of the document data 21 is manually inputted as is to the transmitting device 1.1a, but after inverting it in units of "1" bit, the transmitting device 1, la By doing this, the pixel arrangement of the cell can be changed appropriately even if the bits with the value “θ′′ are consecutive in the document data. It is possible to prevent the dithered image synthesized by the above method from causing any strange feeling.

Furthermore, this method can be widely used not only in the above-mentioned case but also in cases where there is a possibility that a certain regular data string may have an undesirable effect on a reproduced image.

Furthermore, in each of the above-mentioned embodiments, the present invention has been explained by taking a data transmission/reception system as an example, but the present invention may be applied to systems other than such a system. If applied to a system that synthesizes and centrally manages personal information, a third party will be able to recognize the data only as a face photo, and only those with a matching encryption key will be able to know the confidential data. It is possible to create a system that is completely safe.

Furthermore, regardless of the presence or absence of an encryption method, if the present invention is applied to facsimile machines used in daily life or other image signal transmission, image-related data and other data can be transmitted all at once. This makes it possible to unify the transmission process.

Furthermore, it is obvious that the mixed data is not limited to mere characters, but may also be data such as audio signals or image signals.

(Effects of the Invention) As explained above, according to the present invention, a large amount of data can be mixed into image data, and the mixed data and its existence can be prevented from appearing on the playback screen. , this can prevent the image quality from deteriorating.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of a data transmission/reception system to which an embodiment of the data synthesis method according to the present invention is applied, and FIG. 2 is a schematic diagram showing an example of a cell used in the embodiment.
31st! I (A) and (B) are the encoding and
FIG. 4 is a flowchart for explaining the decoding procedure, FIG. 4 is a table for explaining combination examples for each concentration level m in the same embodiment, and FIG. 5 is a flow chart for explaining the amount of information that can be synthesized into each cell in the same embodiment. [obtained by encoding] A table showing the amount of information contained in a pixel, etc.; FIG. 6 is a table showing the input/output bit rate of the transmitting device used in the same example;
FIG. 7 is a block diagram showing an example of applying the present invention to encrypted communication means, FIG. 8 is a flowchart showing an example of a cell determination procedure that can be used in the present invention, and FIG. FIG. 2 is a schematic diagram for explaining the processing procedure shown in FIG. 1... Transmitting device, 2... Receiving device, 3... Image capturing section, 4... Quantizing section, 5... Density gradation section, 6...
... Matrix selection section, 7... Encoding section, 8...
Dictionary section, 9... Matrix detection section, 10... Dictionary section, 11... Data storage section.

Claims (1)

[Claims] A data synthesis method characterized by determining a dither matrix based on multivalued data obtained by quantizing density information of an original image and data to be mixed.