JPH10257300A - Picture processing method and picture processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To precisely read multiplexed auxiliary information by changing the distance from an original point at the time of polar coordinate-displaying a frequency component obtained by executing discrete Fourier transformation on a picture block and repetitively multiplexing auxiliary information and identification information showing the termination of multiplexing for one block. SOLUTION: An original picture 1 is divided into blocks larger than the block size used in general non-reversible compression in a block division part 11 and discrete Fourier transformation is executed at every block in a frequency transformation part 12. A frequency component matrix is obtained and the amplitude of the frequency component is normalized in a frequency component normalizing part 13. The arbitrary component value in the normalized component matrix is changed at a change ratio 18. Auxiliary information 2 is repetitively multiplexed in a multiplex processing part 14. The multiplexed frequency component matrix is inversely normalized and inverse discrete Fourier transformation is executed. Contrast adjustment is executed in a picture contrast adjusting part 16 so that the picture after inverse conversion does not exceed the definition area of a picture element value and a multiplexed picture 3 is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image processing method and apparatus, and more particularly, to multiplexing another sub-information with a digital image so as to prevent information from being perceived by human perception. The present invention relates to a method and apparatus for performing multiplexing and reading out multiplexed sub-information from a digital image.

[0002]

2. Description of the Related Art Today, such information multiplexing and reading techniques are widely used in a system for protecting copyrights of digital information contents and preventing unauthorized duplication by multiplexing copyright information and user IDs into information contents. Have been.

However, the conventional technique has a problem that multiplexed sub-information is easily erased by performing partial editing, lossy compression, or the like of a digital image. In particular, in irreversible compression, sub-information is more likely to be erased by deleting pixel information more greatly in a flat region than in a complicated region of an image, so if irreversible compression is performed on an image with many flat portions, However, there is a problem that reading of the sub information fails. In addition, there is a problem that multiplexing is difficult because the flat portion is relatively easily perceived by humans.

[0004]

SUMMARY OF THE INVENTION It is an object of the present invention to partially edit or compress an image, particularly in the case of an image having a lot of flat portions, which is a problem of the above-mentioned prior art which is not perceived by humans. For lossless compression, multiplexed sub-information can be read correctly, and the performance of reading multiplexed information can be improved, and at the same time, degradation of image quality due to multiplexing can be reduced. To provide an image processing method and apparatus.

[0005]

According to the present invention, an image is subdivided into blocks larger than, for example, an 8 × 8 block size used in general lossy compression, and frequency components obtained by performing a discrete Fourier transform on the blocks are displayed in polar coordinates. By changing the size (distance from the origin) when
By repeatedly multiplexing the block with the sub-information and the identification information indicating that the block has been multiplexed, the sub-information can be correctly read even with lossy compression. In addition, by performing multiplexing and sub-information reading after normalizing the frequency component to a range up to a predetermined value, image processing that is weaker than a complex area in a flat part is performed, and deterioration of image quality due to multiplexing is suppressed, At the same time, resistance to contrast changes is provided.
Further, the larger the value of the frequency component to be changed is, the larger the change amount of the frequency component is (the smaller the smaller the value is, the smaller the change amount), thereby further suppressing the deterioration of the image quality. Further, when subdividing into blocks, in order to correspond to an arbitrary image size, an image area smaller than one block is filled up with an average value of its pixel values or a portion which is not enough by repeating a line-symmetrical figure, thereby reducing the size to one. Treat it as a block. Further, for each block, the reliability of the information read from the block is obtained, and the weight of the information for each block is added, and then the sub-information read from the entire image is formed. The correct sub-information can be read even if the editing is performed erroneously or an image having many flat portions is irreversibly compressed.

[0006]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing an overall configuration and a processing flow of an embodiment of an information multiplexing apparatus according to the present invention. The information multiplexing device 10 outputs a multiplexed image 3 with the original image 1, the sub-information 2 to be multiplexed, and the multiplexing parameter 7 for specifying a multiplexing method. FIG. 3 shows an example of the configuration of the multiplexing parameter 7, in which a block size 71 for specifying the size of a block for subdividing the original image is shown.
(This is assumed to be larger than the normal 8 × 8 size in order to sufficiently provide resistance to irreversible compression.) Multiplexing which determines the difficulty of erasing multiplexed sub information and the degree of image deterioration due to multiplexing processing A multiplexing strength parameter 72, a multiplexing key 73 for ensuring security against intentional deletion of sub-information from a multiplexed image, and a block for specifying the number of times to repeatedly multiplex sub-information per block The number of repetitions is 74.

In the information multiplexing apparatus 10, first, an original image (digital image) 1 is divided by a block division unit 11 into blocks each having a size designated by a block size 71 which is one of the elements in a multiplexing parameter 7. Subdivide into
When the original image 1 is subdivided into blocks, if the area at the end of the image is less than the block size 71, the missing part is filled by a predetermined method to form one block. This will be described later. Next, the frequency converter 12
, A discrete Fourier transform is performed for each subdivided block to obtain a frequency component matrix. Next, the frequency component normalization unit 13 normalizes the amplitude of the frequency component to obtain a normalized frequency component matrix. In addition, the frequency component normalizing unit 13 sets a ratio representing the extent to which the frequency has been enlarged / reduced at the time of normalization as the change ratio 18 and the inverse transform unit 15.
Send to Next, the multiplexing processing unit 14 changes some component values in the normalized frequency component matrix so that the sub-information 2 is changed by the intra-block repetition count 74 which is one of the elements of the multiplexing parameter 7. Repeat multiplexing. The degree of change of the component value is specified by a multiplexing intensity parameter 72 which is one of the elements of the multiplexing parameter 70. The selection of the frequency component to be subjected to the multiplexing process is performed by selecting a multiplex key 73 which is one of the elements of the multiplex parameter
Is determined by Next, in the inverse transform unit 15, the sub-information multiplexed frequency component matrix is subjected to inverse normalization and inverse discrete Fourier transform. Finally, the image contrast adjusting unit 16
Then, if necessary, perform a pixel value range over avoiding process,
A multiplexed image 13 is obtained.

FIG. 2 is a diagram showing the overall configuration and processing flow of an embodiment of the information reading apparatus according to the present invention.
To read the sub information, a read target image 5 such as the multiplexed image 3 and a multiplexing parameter 7 are required. Here, the configuration of the multiplexing parameter 7 is as shown in FIG.

The information reading device 20 first subdivides the image 5 to be read into blocks in the block dividing unit 21 similarly to the information multiplexing device 10 in FIG. If the end portion of the image 5 to be read out of the subdivided blocks is less than the block size 71 of the multiplexing parameter 7, the missing portion is filled to one block. The block division unit 21 calculates the area ratio between the portion of the image 5 to be read and the block size 71 for each block, and provides the area ratio 25 for each block to the sub-information reading processing unit 24. Naturally, when the block to be subdivided is filled with the portion of the image 5 to be read, the value of the area ratio per block 25 is 1. Next, the frequency converter 22
Then, a discrete Fourier transform is performed for each block to obtain a frequency component matrix. Next, the frequency component normalization unit 23 normalizes the amplitude of the frequency component to obtain a normalized frequency component matrix. In the frequency component normalizing section 23, 1 in FIG.
There is no need to determine the change ratio as in 3. The sub-information reading processing unit 24 determines the frequency component to be read using the multiplexing key 73 of the multiplexing parameter 7 in the same manner as the multiplexing processing unit 14 of FIG. 1, and performs multiplexing in the reverse procedure of the multiplexing process. And retrieve the information. Then, a block weight indicating the reliability of the information read from each block is calculated, and the read sub-information 6 is obtained by reflecting the block weight on the information read from each block.

If the image 5 to be read is an image obtained by multiplexing the sub-information 2 by the information multiplexing device 10 using the multiplexing parameter 7, the read sub-information 6 matches the sub-information 2, and It can be read. If the image 5 to be read is not an image multiplexed with the multiplexing parameter 7 or is a multiplexed image 3 that has been severely degraded, the read sub-information 6 does not match the sub-information 2 and reading has failed. become.

An example of the configuration of each unit shown in FIGS. 1 and 2 and the contents of the processing will be described in detail below. FIGS. 4 and 5 are diagrams showing the contents of the processing of the two embodiments in the block division unit 11 of FIG. 1 and the block division unit 21 of FIG. The block division units 11 and 21 subdivide the image (the original image 1 or the image 5 to be read) into blocks 401 having a block size 71 of the multiplexing parameter 70. At this time, for the image region less than one block at the end of the image, in the case of the embodiment of FIG. 4, the average pixel value of the image region is calculated, and the portion that is not enough to make one block is averaged. This is a block filled with pixels. Further, in the case of the embodiment shown in FIG.
Try to fill the blocks. Note that the information reading side block dividing unit 21 calculates the ratio (area ratio) of the image to be divided in the block 401 for each block 401, and sets the area ratio per block to 25.

FIG. 6 is a diagram showing the processing of the frequency converter 12 of FIG. 1 and the frequency converter 22 of FIG. The frequency converters 12 and 22 perform a discrete Fourier transform using the image block 401 as an input to obtain a frequency component matrix 501. This is performed for each block subdivided by the block division units 11 and 21.

FIGS. 7 and 8 are diagrams showing the contents of the processing of the two embodiments in the frequency component normalizing section 13 of FIG. 1 and the frequency component normalizing section 23 of FIG. FIG. 7 shows a ratio of enlarging the maximum amplitude component value in the low frequency region of the frequency component matrix 501 (FIG. 6) to a normalized value a (a is a sufficiently large value) (this is a change ratio). Indicates that a normalized frequency component matrix 601 is obtained by enlarging all frequency components. FIG. 8 shows the average value of the absolute values of the component values in the low frequency region of the frequency component matrix 501, and the ratio (change ratio) of expanding this average value to a normalized value a (a is a sufficiently large value). , All frequency components are enlarged to obtain a normalized frequency component matrix 601. In the frequency component normalizing section 13 on the information multiplexing side, the change ratio at this time is transmitted to the inverse transform section 15. Although FIGS. 7 and 8 show analog waveforms for easy understanding, the frequency component matrices 501 and 601 are of course digital values.

FIG. 9 is a diagram showing a detailed configuration and a processing flow of the multiplex processing section 14 of FIG. Multiplex processing unit 1
First, in the identification information adding unit 144, the sub information 2
Is added to the sub-information 2 with identification information indicating that multiplexing has been completed (a bit pattern in which bit 1 and bit 0 appear at the same frequency. For example, an 8-bit identification code such as 10101010) is added. The multiplexing information is 145. On the other hand, the random number generator 140
A random number sequence 141 is generated by using the multiplexing key 73 of. Then, the multiplexing component determination unit 142 determines to which component in the normalized frequency component matrix 601 the multiplexing process is to be performed using the random number sequence 141, and divides the selected frequency component into the multiplexing component 143. And Multiplexed bit determination unit 14
In 6, one bit is sequentially extracted from the top bit of the multiplex information 145 toward the lower bit, and
47. If the lowest bit of the multiplex information 145 has been reached, it is returned to the first bit. The frequency component changing section 148 receives the multiplexing component 143, the multiplexing bit 147 and the multiplexing intensity parameter 72 of the multiplexing parameter 7, and changes the component value of the multiplexing component 143 to obtain the multiplexed component value 149. obtain. That is, the multiplexed component 143 in the normalized frequency component matrix 601 is changed to the multiplexed component 149.

The above processing is repeated by (the bit length of the multiplexing information 145) × (the number of repetitions 74 in the block). That is, one normalized frequency component matrix 60
1, the multiplexing information 145 is multiplexed by repeating the number of times 73 in the block. In FIG.
The portion enclosed by the dotted line indicates that (the bit length of the multiplexed information) × (intra-block repetition processing is performed per block. In this manner, the normalized frequency component row in which all the bits have been multiplexed has been completed). 601 is a multiplexed frequency component matrix 801. The multiplex processing unit 14 performs this for each block.

FIG. 10 is a diagram showing the detailed configuration and processing flow of the frequency changing unit 148 of FIG. The frequency changing unit 148 first converts the value of the multiplexing component 143, which is a complex number represented in the form of a + bi, into a polar coordinate display converting unit 148.
1, the size (r) 1482 and the angle (θ) 148
3 is a polar coordinate display 143 ′ in the form of r · exp (iθ). Next, in the r change processing unit 1484, the size (r) 1482 of the polar coordinate display 143 ', the multiplexing intensity parameter 72 and the multiplexing bit 147 are input, and the size (r) 1482 is changed to the size (r'). Change to 1485.
Finally, in the Gaussian plane display conversion unit 1486, r ′ ·
The complex number of exp (iθ) is converted into the form of x ′ + iy ′, which is used as a multiplexed component 149.

Here, the processing of the r change processing section 1484 will be described in more detail. The r change processing unit 1484 inputs the multiplexed bits 147, the multiplexed intensity parameter 72, and the size (r) 1482 of the multiplexed component 143 'displayed in polar coordinates, and changes it to the size (r') 1485. In particular,
As shown in FIG. 15, the frequency component values are sequentially assigned to h, 2 from 0 using the multiplexing intensity parameter 72 (h in FIG. 15) so that the larger the value of r is, the larger the change amount is.
.. are divided into widths of h, 3h,..., and a frequency component value belonging to less than half of the width represents bit 0 and a frequency component belonging to more than half represents bit 1 for each width.
Here, the size (r ′) obtained by multiplexing the multiplexed bits on the size (r) is represented by the size (r) and the size of the multiplexed bit that minimizes the change amount of the size (r). It is obtained by changing to an intermediate value in the area indicating the bit value.

FIG. 11 is a diagram showing the contents of the processing of the inverse converter 15 in FIG. Although FIG. 11 shows an analog waveform for easy understanding, it is actually a digital value. The inverse transform unit 15 converts the multiplexed frequency component matrix 801 obtained by the multiplex processing unit 14 into the frequency component normalizing unit 1
3, the frequency conversion unit 1
Original frequency component matrix 501 generated in step 22 (FIG. 6)
A frequency component matrix 1001 having substantially the same amplitude as the above is obtained, and this is subjected to an inverse Fourier transform to obtain a processing process image 1002.

In the image contrast adjusting section 16 shown in FIG.
If the pixel value in the processing image 1002 obtained by the inverse transform unit 15 is the domain of the pixel value (0 for a 8-bit grayscale image)
２５255), the contrast of the pixel values of the in-process image 1002 is reduced at a rate such that the pixel value that protrudes the largest just becomes the minimum value (or the maximum value) of the defined area. Is corrected so that the pixel value of the image falls within the defined area, and the multiplexed image 3 is obtained.

FIG. 12 shows the sub-information reading processing section 2 of FIG.
FIG. 4 is a diagram showing a detailed configuration and a processing flow of No. 4; In the sub information reading processing unit 24, first, the random number generator 240 generates a random number sequence 241 using the multiplexing key 73 of the multiplexing parameter 7. Next, the multiplexed component determination unit 242
, The normalized frequency component matrix 6 of the multiplexed image
01 ′ is a random number sequence 2 for which component to perform multiplexing processing.
41, and the selected frequency component is set as the multiplexed component 243. The bit value reading section 244 reads the multiplexed bit value from the multiplexing intensity parameter 72 and the multiplexing component 243 in the multiplexing parameter 7 to obtain a read bit value 245. These processes are repeated by the number of repetitions 74 in the block of the multiplexing parameter 7,
All the bit values multiplexed in one normalized frequency component matrix 601 'are read. This operation is performed on all the normalized frequency component matrices 601 ′,
The read bit value 245 extracted for each normalized frequency component matrix 601 'is accumulated in the block-by-block bit value buffer 246.

When all the bit value reading operations have been completed and all the read bit values 245 of all the blocks have been stored in the block-by-block bit value buffer 246, the block weight determining unit 247 sets the block-by-block bit value buffer 246. , The block weight 248 indicating the reliability of the read bit value read from each block is determined. Thereafter, in the read sub-information determination unit 249, the read sub-information 6 from the entire read target image 5 (multiplexed image 3) is constructed from the contents of the block-specific bit value buffer 246 and the block weight 248, and The reading process ends.

Here, the bit value reading section 244 receives the multiplexed component 243 and the multiplexed intensity parameter 72 as inputs and determines a read bit value 245. In this bit value reading unit 244, the r change processing unit 1484 of FIG.
Similarly to the above, the multiplexed component 243 is converted into a multiplexed component expressed in the form of a magnitude (r) and an angle (θ) expressed in polar coordinates, and the value of the magnitude (r) is converted into a rule shown in FIG. Reads 1-bit information.

FIG. 13 is a block diagram showing the block weight determining unit 2 shown in FIG.
FIG. 47 is a diagram showing a detailed configuration and processing of 47. In the block weight determination unit 247, first, in the block-by-block read multiplexing information generation unit 2471, the bit string of the read bit value 245 accumulated in the block-by-block bit value buffer 246 and the number of repetitions 74 of the multiplexing parameter 7 in the block are calculated. , And generates the read multiplex information 2472 for each block having the same bit length as the bit length of the multiplex information. Specifically, the decision is made for each bit using a majority decision method or the like. The block weight calculator 2473 calculates the reliability of the read multiplex information 2472 for each block. The original multiplexed information is composed of sub-width information and identification information. If the identification information is always a fixed-length bit string, as shown in FIG. 14, the identification information and the identification information of the multiplexed information read for each block are used. By comparing the corresponding parts, a different bit number d is obtained. Here, if d is large, it is considered that the reliability of the multiplexed information 2472 read for each block is low. The area ratio 25 per block is a value smaller than 1 for a block in which, for example, padding has been performed with average pixels in the block division unit 21 in FIG. Therefore, when the block area ratio is smaller than 1, it is considered that the reliability of the multiplexed information read for each block is low.

From the above, for example, the block weight 248 is determined for each block simply by the following weight calculation formula.

[0025]

(Equation 1)

The read sub-information determination unit 249 of FIG. 12 adds a corresponding block weight 248 to the column of read bit values 245 stored in the bit value buffer To determine the read sub information 6. For example, by using a weighted majority decision method for each bit position of the read sub-information, the read sub-information 6 can be determined in a form reflecting the block weight 248. Further, the read sub-information 6 can also be configured by performing error correction using the majority decision method. As a result, information from a block in which multiplexed information has been broken is not relied on, and read sub-information 6 is determined based on information on a block considered to remain relatively correctly, so that partial editing, It is possible to minimize the influence of a portion where multiplexed information due to lossy compression is easily broken, such as a flat portion of an image. Further, by using the area ratio per block 25, the multiplexing process can be performed on the entire area region.

[0027]

As described above, according to the present invention,
Even when the image is partially edited or an image with many flat parts is irreversibly compressed, using block weights enables
The correctly multiplexed sub information can be read. In addition, for an image area that is less than one block at the end of the image, the missing part is filled up with an average pixel or a repetition of a line-symmetric figure, and the area ratio is reflected in the block weight. And multiplexing processing for all image areas. Furthermore, by multiplexing / embedding the frequency component after always normalizing it to a predetermined value, it is possible to realize a resistance to a change in the contrast of an image and a reduction in image quality deterioration due to the multiplexing.

Further, since the sub-information is closed and multiplexed for each image block, a block start point is searched for a partial cutout of an image if the cutout area includes one or more blocks. This makes it possible to extract sub-information from the cut-out area. Further, the present invention can be applied to a color image by being carried out for each component of the color image, and also applicable to a moving image.

By using the present invention in a copyright protection system or the like, it is possible to make sub-information more difficult to erase with higher accuracy even in the case of partial editing or irreversible compression of an image, as compared with the conventional method. The degradation of image quality due to multiplexing can be reduced more than in the method.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration and a processing flow of an embodiment of an information multiplexing apparatus according to the present invention.

FIG. 2 is a diagram showing a configuration and a processing flow of an embodiment of an information reading apparatus according to the present invention.

FIG. 3 is a diagram illustrating a configuration example of a multiplexing parameter used in the present invention.

FIG. 4 is a diagram showing the contents of processing of a first embodiment of a block division unit.

FIG. 5 is a diagram illustrating the content of processing of a second embodiment of the block division unit.

FIG. 6 is a diagram illustrating processing of a frequency conversion unit.

FIG. 7 is a diagram illustrating a process of a first embodiment of a frequency component normalization unit.

FIG. 8 is a diagram illustrating a process of a second embodiment of the frequency component normalizing unit.

FIG. 9 is a diagram illustrating a configuration example of a multiplexing processing unit and a processing flow.

FIG. 10 is a diagram showing a detailed configuration example and a processing flow of a frequency component changing unit in FIG. 9;

FIG. 11 is a diagram illustrating processing of an inverse transform unit.

FIG. 12 is a diagram illustrating a configuration example of a sub information reading processing unit and a processing flow.

13 is a diagram illustrating a detailed configuration example and a processing flow of a block weight determination unit in FIG. 12;

FIG. 14 is a diagram illustrating a method of obtaining a different bit number d between identification information and identification information of read multiplexed information for each block.

FIG. 15 is a diagram illustrating a method of changing a frequency component and a reading method according to the present invention.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 original image 2 sub-information 3 multiplexed image 5 image to be read 7 multiplexing parameter 10 information multiplexing device 11 block division unit 12 frequency conversion unit 13 frequency component normalization unit 14 multiplex processing unit 15 inverse conversion unit 16 image contrast Adjustment unit 21 block division unit 22 frequency conversion unit 23 frequency component normalization unit 24 sub-information reading processing unit

──────────────────────────────────────────────────の Continued on front page (51) Int.Cl. ⁶ Identification code FI H04N 5/91

Claims (16)

[Claims] 1. An image processing method for multiplexing another piece of sub-information in a digital image and reading the multiplexed sub-information, dividing the digital image into blocks, performing frequency conversion for each block, and An image processing method, comprising: multiplexing sub-information by changing a value; and reading the multiplexed sub-information from the changed frequency component. 2. The image processing method according to claim 1, wherein
When the digital image is subdivided into blocks, an area at an end of the image that is smaller than a predetermined block size is configured to fill one part with an insufficient average pixel value of the area to constitute one block. Image processing method. 3. The image processing method according to claim 1, wherein
When a digital image is subdivided into blocks, one block is formed by repeating a line-symmetrical figure of the region at an end portion of the image smaller than a predetermined block size so as to fill the block. Image processing method. 4. The image processing method according to claim 1, wherein when changing the value of the frequency component and when reading the sub-information, the amplitude range of the frequency component is always determined for any block. A multiplexing process and a reading process after normalizing to a normalized value. 5. The image processing method according to claim 1, wherein a discrete Fourier transform is used as a frequency transformation, and the size (distance from the origin) of the frequency component value when displayed in polar coordinates is changed. An image processing method characterized by multiplexing. 6. The image processing method according to claim 1, wherein when multiplexing the sub-information for each block, the sub-information is repeatedly included in one block. 7. The image processing method according to claim 1, wherein when multiplexing the sub-information for each block, the sub-information is multiplexed by adding identification information indicating that the multiplexing has been completed. Image processing method. 8. The image processing method according to claim 1, wherein when multiplexing information by changing the frequency component, the larger the frequency component to be changed, the larger the change amount. An image processing method characterized in that the smaller the value, the smaller the change amount. 9. The image processing method according to claim 1, wherein after the frequency information is changed to multiplex the sub-information, an inverse frequency conversion process is performed, and the pixel value exceeds the domain of the pixel value. An image processing method characterized by performing a non-process. 10. The image processing method according to claim 9, wherein the contrast of the image block after the inverse frequency conversion is reduced as the processing in which the pixel value does not exceed the defined range of the pixel value. 11. The image processing method according to claim 1, wherein at the time of reading the sub-information, the reliability of the information read from within the block is obtained for each block, and the reliability is used as a weight from the entire image. An image processing method comprising: determining read sub-information of the image. 12. The image processing method according to claim 11, wherein when reading the sub-information, a difference between the multiplexed identification information equivalent portion and the original identification information is determined for each block from within the block. An image processing method characterized by determining the reliability of read information. 13. The image processing method according to claim 1, wherein when reading the sub-information, for each block, read the information read from within the block by using a majority decision method to form the sub-information. Characteristic image processing method. 14. The image processing method according to claim 13, wherein when reading the sub-information, the reading sub-information is determined from the entire image by using a majority decision method in consideration of the weight of each block. Characteristic image processing method. 15. A means for inputting an image, sub-information to be multiplexed on the image and multiplexing parameters, divides the image into blocks, performs frequency conversion for each block, and normalizes the frequency components. Means for multiplexing sub-information by changing the value of the normalized frequency component,
Means for performing an inverse frequency conversion of the multiplexed frequency component,
Means for preventing the image after the inverse frequency conversion from exceeding the defined range of pixel values, wherein the sub-information is multiplexed in the image without visually affecting the image. Processing equipment. 16. A means for inputting a multiplexed image and multiplexing parameters, dividing the image into blocks, a means for performing frequency conversion for each block, a means for normalizing the frequency component, and a means for normalizing the frequency component. Means for reading multiplexed sub-information from the frequency components obtained, and correctly reads the multiplexed sub-information even if the multiplexed image is degraded.