JP3682864B2 - How to embed invisible information - Google Patents

Description

[0001]
[Field of the Invention]
The present invention relates to a document creating apparatus for forming a document with pixels such as a digital copying machine, a digital printer, an ink jet printer, a printing machine, and the like, and in particular, information different from visually perceived document information is included in a part of the document. The present invention relates to a document creating apparatus characterized by embedding in a state that cannot be perceived by the naked eye, and is characterized by embedding the other information in the document at high speed.
[0002]
[Prior art]
Conventionally, in the field of document creation, for the purpose of increasing the amount of information or concealing it to a third party, information other than the document, such as text and characters, is added and mixed in the image. ing.
[0003]
For example, Japanese Patent Publication No. 2958396 describes a method of adding such information to a document in a line screen or a dot screen. However, with this method, information cannot be embedded where no line or dot screen is printed. In this patent, it is stated that it is only necessary to embed information by performing screen printing to such an extent that it does not become unobtrusive even in such a portion where there is no line screen or dot screen. It is often not allowed to print inside. In addition, it is difficult to read information embedded in a screen that does not obstruct the screen.
[0004]
In the book “Basics of Digital Watermarking” by Kokoo Matsui (Morikita Publishing), there are various methods for embedding information in a document. It is described that various methods have been proposed and implemented for images with relatively redundancy such as screen printing, and even for images with little redundancy such as text only. A method of changing the character pitch, a method of changing the character width, a method of changing the inclination of the character, and the like are described.
[0005]
In recent document copying machines such as digital copying machines, digital printers, ink jet printers, and printing machines, the resolution is greatly improved and almost no inferiority compared to offset printing. However, it is difficult to embed information by this method because adding a character pitch, a character width, or a character inclination in a character string directly affects a decrease in print quality. Also in this case, it is difficult to read out the information embedded by the character pitch, the character width, the character inclination or the like unless the information about the original character is known.
[0006]
In 1999, Inspectron Co., Ltd. presented a method for printing information on an image with a product name “stealth dot” and adding information to the image without degrading the print quality. However, in the proposed method, the dot diameter to be printed is typically 0.1 mm, and this size of the dot is visible. If the information to be recorded by this stealth dot is several bytes, the impression that the image has been specially soiled by embedding this information when viewing the image may be escaped. When it is desired to embed a larger amount of information, for example, several hundred bytes, with dots, the information recorded portion may be seen as a dark portion, which may cause the quality of the image to deteriorate.
[0007]
On the other hand, it is well known that there is a problem that the reliability of recording deteriorates when trying to embed small information that is difficult to see. For example, in the case of a laser beam printer, it becomes difficult to reliably develop and transfer toner. It becomes difficult to distinguish from particles, which are called “fogging”, in which toner is scattered to portions other than the recording portion. Even in the case of inkjet, printing of isolated minute dots is absorbed by the recording paper and the recording density is extremely lowered. Further, in addition to the ink particles originally used for recording, which are called ink satellites, it becomes difficult to distinguish the fine particles generated when the ink particles are separated from the recording paper. However, by embedding in the document with regularity dots (dots) that are larger than the above-mentioned fog and inadvertently adhering to the recording paper and that are difficult to see, they are visible to the eyes. Patterns that are difficult but meaningful as information can be formed in a document.
[0008]
However, in order to embed information formed from such minute dots in a document, it is necessary to find a place where nothing is printed and there is a sufficient area for recording on each page in the document. , It is necessary to embed information in that place. In a small document of several pages, the processing time is not so noticeable even if a blank portion is obtained for each page. However, in a document of tens or hundreds of pages, the processing time for obtaining the blank portion is added. Therefore, there is a problem that the processing time of the document is remarkably reduced.
[0009]
[Problems to be solved by the invention]
The present invention is intended to solve the above-mentioned conventional problems, and provides a method for embedding a large amount of additional information on each page in a document in a state that cannot be perceived by the naked eye without significantly reducing the processing speed. For the purpose.
[0010]
[Means for Solving the Problems]
The present invention is a document creation apparatus including a first digital image output unit that visualizes information to be printed on a document on a sheet, and a second digital image output unit that records information different from the information to be printed. The first digital image output means extracts a blank portion common to a plurality of pages of the document, and the second digital image output means is a pixel having a size that is substantially indistinguishable with the naked eye and with the naked eye. This can be realized by a method of embedding invisible information in a sheet, wherein an image having an unrecognizable density is recorded in a blank area common to a plurality of pages extracted by the first digital image output means of the document .
[0011]
DETAILED DESCRIPTION OF THE INVENTION
The human visual acuity is set to 1.0 with a limit that can be perceived as two points when light emitted from two different points forms an image on the retina at an angle of 1 minute (1/60 degrees). If this is expanded, it means that a person with a visual acuity of 1.0 can finally recognize two points 75 μm apart at a clear visual distance of 25 cm. That is, a person with a visual acuity of 1.0 cannot recognize a point or gap that is less than 75 μm in the clear vision distance.
[0012]
However, even if it is a point of 75 μm or less, if there are a large number of such points in the unit area, the individual points cannot be distinguished, but they are recognized as a change in density as a whole. It is said that the difference in density with the human eye is about 0.01 in terms of optical density. Actually, in a digital image forming apparatus that expresses an image with binary values of black and white, such as a printer or a digital copying machine, an image such as a photograph is expressed in 8 × 8 units, that is, 65 halftones. Can do.
[0013]
Therefore, the unit of the image is densely arranged with one unit being an image constituting unit of the digital image apparatus, in which one pixel is less than 75 μm and the black pixel is present at the center of the 8 × 8 white pixel group. Even if it is done, it is not perceived by the human eye with a visual acuity of 1.0.
[0014]
This is one means for preventing the density difference from being recognized with the naked eye very strictly, but it is not always necessary to prevent the density difference from being recognized in this way. For example, in general, even if there is some density in the peripheral part of a digital image, it is not obstructive to the naked eye. In general, in the peripheral portion or the like, if the density is 0.1 or less, the digital image is not disturbed. For that purpose, even if halftone is expressed in units of about 3 × 3, there is no problem.
[0015]
FIG. 1 is a diagram illustrating the principle of the present invention. Reference numeral 1 denotes one page of a document output from a digital image forming apparatus such as a printer or a digital copying machine. Reference numeral 2 denotes a character portion in one page of the document. Reference numeral 3 denotes an image portion such as graphics or a photograph in one page of the document. Reference numeral 4 is a blank portion in one page of the document, and 5 is information different from the perceived information embedded in a part of the blank portion. 6 is an enlarged view of a part of the information, and shows a state in which each significant block is divided into blocks. 7 shows the details of the significant block. One significant block is further divided into six significant sub-blocks. 8 is a further enlargement of the significant sub-block, and one sub-block is composed of 8 × 8 pixels. 9, 9 ′, 9 ″ are information of “1” embedded in the significant sub-block 7. The specific pixel at the center of the 8 × 8 pixels is used as a representative point of the sub-block, and Whether the sub-block is “1” or “0” is determined depending on whether or not the information of “1” is embedded.
[0016]
Here, as an example, the significant block 6 is divided into six significant sub-blocks, but the number of divisions is not limited to six. Since the Japanese braille notation system expresses a single kana character with six dots as one block, the notation system is borrowed for an example of embedding information for easy understanding.
[0017]
In the above description, an example in which one sub-block is composed of 8 × 8 pixels is described. However, the pixel configuration is generally the significant block when all the sub-blocks are “1”. It is desirable to select the number of constituent pixels so that the image density when viewed as an image is 0.1 or less. In other words, even if the representative points of the sub-block cannot be determined with the naked eye, it is not preferable that the representative points are densely present, and it is not preferable that the sub-block is visible to the naked eye. Preferably, the representative point is not perceived if the image density is 0.01 or less.
[0018]
FIG. 2 is a schematic diagram showing an example in which specific information is written in the significant block 7 including the six significant sub-blocks 8 described above. In order to improve the reliability of the embedded information, the always-printed block 11 in which “1” is always written in the significant block 7 and the number of “1” in the significant block 7 must be an even number. A block 12 for parity check is provided. The remaining four blocks are determined to be “0” or “1” depending on the information to be embedded.
[0019]
If determined in this way, 16 kinds of information can be recorded from the significant block 7. Furthermore, the recorded information always has “1” information in the upper left, and the parity check that the number of “1” s present in the position-dominant block is an even number. Can be confirmed.
[0020]
In this case, it is desirable that the parity check is performed with an even number, and if it is an odd number, if one “1” point in the significant block is not recognized, there is no data there. This is to avoid that.
[0021]
Also, the reason for providing a significant sub-block whose fixed position is always “1” is that the sub-block can be expected to appear at regular intervals so that the presence position of significant block recognition can be easily found. It is to do.
[0022]
FIG. 3 is a schematic diagram showing a method of embedding the storage management information of the print document in a plurality of locations in the print document. In FIG. 1, the storage management information is embedded in one place and 5 in the printed document, but FIG. 3 shows 5a, 5b, 5c, or the storage management information in the horizontal direction in addition to 5. And 5d, 5e, etc. arranged in the vertical direction can be embedded in a plurality of locations.
[0023]
FIG. 4 is a block diagram showing an algorithm for embedding invisible information according to the present invention. Reference numeral 30 denotes a logical description portion of a print page in the document. Reference numeral 31 denotes an image development unit. 32 is a developed image portion of the first page, 32 ′ is a developed image portion of the second page, 32 ″ is a developed image portion of the third page, and so on, and 34 is a developed image portion of the Nth page. Is an integration unit for developed images, 35 is an integration unit for developed images of all pages, 36 is a blank extracting unit for the developed images 35 of all pages, and 37 is information different from the visually perceived image. 38 is an invisible pattern conversion unit, 39 is a synthesis unit, and 40 is a printer.
[0024]
The logical description part 30 is a part expressed by a code called a page description language. Generally, a page image printed by a printer has such a logical description part. The code of the logical description unit 30 is interpreted by the image development unit 31 by a program called an interpreter and is developed into bit map data corresponding to the pattern of the page image to be printed and 1: 1. .. Are stored in the developed image parts 32, 32 ′, 32 ″,... At the same time, the pattern of the page image developed in this way is sent to the developed image integrating part 33, and the integrated value of the page image is taken. Specifically, in an image in which white is represented by “0” and black is represented by “1”, a pattern where a logical sum “or” is taken and black is present throughout the entire page is a black pixel. Conversely, in an image in which white is represented by “1” and black is represented by “0”, the logical product “and” is taken to obtain a similar pattern.
[0025]
The blank extraction unit 36 extracts a blank part from the image patterns of all pages obtained in this way. Ideally, a plurality of rectangular blank portions are extracted from the printed page image. On the other hand, information 37 different from the printed image, that is, the visually perceived image is prepared. For this information, various attributes such as detailed attributes of the page to be printed, file storage location and author data, correction history of the page can be applied. Information 37 different from the perceived image is converted into an invisible pattern as shown in FIG. The converted invisible pattern is synthesized and output by the synthesis unit 39 to the blank portion extracted by the blank extraction unit 36 and output to the printer 40.
[0026]
Further, by embedding such an invisible pattern not only in one place of the digital image but also in a plurality of places, even if the invisible pattern in one place cannot be read, the pattern in another place can be read. Reliability can be improved by being able to do so.
[0027]
FIG. 5 is a block diagram showing an embodiment of an algorithm for reading the storage management information embedded according to the present invention, which is invisible or invisible unless it is watched. Reference numeral 41 denotes a scanner, 42 denotes an isolated point extraction unit, and 43 denotes a dominant block lump detection unit. Here, in a plurality of significant block clusters, 44 is a pattern decoding unit, 45 is a data decoding unit, and 46 is a data string generating unit. 44 is a pattern decoding unit, 45 is a data decoding unit, and 46 is a plurality of data string generating units as required. Reference numeral 47 denotes a mutual comparison unit for each data string, and 48 denotes a data determination unit.
[0028]
The printed page embedded with the storage management information is read by the scanner 41. As is apparent from the description of FIG. 2, the so invisible information is formed in a pattern composed of isolated points in the substantially central portion of the significant sub-blocks of 8 × 8, there is a regularity. The block of significant blocks is detected using this regularity. There are various detection methods, but the methodologies are not the gist of the present invention, so they are omitted.
[0029]
Each significant block performs a parity check to check whether the number of “1” s in the significant block is an even number. If the number of “1” s in the parity check is an odd number, it is reserved as indefinite data, and a data string including the data to be reserved is created.
[0030]
Next, when the uncertain data is generated, according to other needs, or for the purpose of obtaining the reliability of the read data, the same processing is performed on a plurality of detected blocks of significant blocks. And each data string can be created. In that case, it is possible to determine the reserved data in the process of comparing a plurality of detected data strings. The storage management information determined in this way is sent to a storage management instruction section (not shown) , and a corresponding document is searched from the storage (not shown) , and a reprint instruction is issued. Can be given.
[0031]
FIG. 5 shows only one embodiment, and other methods for increasing reliability in the category of software are also known. One is a method to remove noise by superimposing patterns at multiple positions, a method to determine whether there is a dot at a position predicted using regularity, and storage management information to fixed-length data A method for checking whether or not such data is damaged, a parity check for the entire embedded storage management information, and so on. In the present invention, since the method of reading data with high reliability is not the main point, detailed discussion thereof will not be made.
[0032]
【The invention's effect】
According to the present invention, it is possible to provide a method of embedding invisible information in such a state that a large amount of additional information cannot be perceived by the naked eye on each page in a document without significantly reducing the processing speed.
[Brief description of the drawings]
FIG. 1 is a principle explanatory view showing the principle of the present invention.
FIG. 2 is a schematic diagram showing an example in which specific information is written in a significant block composed of the six significant sub-blocks described above.
FIG. 3 is a schematic diagram illustrating a method for embedding the storage management information of the print document in a plurality of locations in the print document.
FIG. 4 is a block diagram illustrating an algorithm for embedding invisible information according to the present invention.
FIG. 5 is a block diagram showing an embodiment of an algorithm for reading storage management information embedded according to the present invention, which is invisible or invisible unless watched.
[Explanation of symbols]
1 is one page of a document output from the digital image forming apparatus, 2 is a character portion of the document, 3 is an image portion of the document, 4 is a blank portion of the document, and 5 is embedded in a part of the blank portion. Information different from the perceived information, 6 is an enlarged view of a part of the information, 7 is a detailed view of the significant block, 8 is a further enlarged view of the significant sub-block, 9 , 9 ′, 9 ″ are information of “1” embedded in the significant sub-block 7.

Claims (3)

A document creating apparatus comprising: first digital image output means for visualizing information to be printed on a document on paper ; and second digital image output means for recording information different from the information to be printed. The first digital image output means extracts a blank area common to a plurality of pages of the document, and the second digital image output means has pixels of a size that is substantially indistinguishable by the naked eye and has a density that cannot be recognized by the naked eye. A method of embedding invisible information in a sheet , wherein an image is recorded in a blank portion common to a plurality of pages extracted by the first digital image output means of the document. The first digital image output means, characterized in that it has a means for developing at least the document to the image data of each print page, and means for holding the integrated calculation result of the image data of the respective print page, A method for embedding invisible information in a sheet according to claim 1. 3. The method of embedding invisible information in a sheet according to claim 2, wherein the result of the integration operation is calculated so that a logical sum is obtained with respect to black pixels.

Images