JP2009017554A - Automatic read system, information embedding device, information embedding program, information recognition device, information recognition program, and automatically read printed matter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of maximizing information recording capacity by embedding a large amount of information in a paper document and forming a dot pattern and a woven pattern minimizing deterioration in picture quality of the paper document. <P>SOLUTION: An information embedding device has the steps of: entering additional information data (S1); converting a pattern-recognizable geographic or physical screen code having characteristics of a screen in combination (S3); entering data of a document (S2); converting image data (S4); arranging the screen code in a matrix form and embedding information (S5); and outputting a print image of a new constituted document and creating automatically read printed matter having information embedded (S6). An information recognition device has the steps of: reading the automatically read printed matter by an image sensor (S'1); detecting positions of respective dots of the screen code (S'2); recognizing multi-bit values (S'3); and registering recognition results in a recording medium (S'4). The availabiltiy of the method is characterized in that a large amount of information can be embedded and there is not a feeling of incompatibility with a paper document. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an automatic reading system for a multifunction machine, an information embedding device, an information embedding program, an information recognition device, an information recognition program, and an automatic reading printed matter.

  In recent years, in order to reduce the cost of storage, operation, and management of paper documents, the transition from paper documents to computerization has been rapidly progressing due to the promotion of national policies. However, there is a general method for registering a paper document as image data with a scanner in order to process from a paper document to digitization. There remains a problem that image data cannot be edited or registered in the database.

  On the other hand, as a typical technique of reading technology, many applications of “character reading device” (see Patent Document 1) and “reading device and reading method” (see Patent Document 2) have been published. In these applications, a method for automatically reading paper documents and paper slips using character recognition OCR technology has been proposed, and the OCR technology can recognize character patterns and convert character codes. Since it does not obtain information or information such as the database format of the slip, it is difficult to fully automate because human intervention is required.

  If you can embed print data in the form of a copy-forgery-inhibited pattern as additional information and print it at the same time as the print data, you can register a digital document with a large amount of information as a recording medium at any time and register it with a computer. There is an advantage that multiple inputs are not required, and there is no need for human intervention for paper documents, and it can be read completely automatically.

  As for a method of embedding information with a tint block, there has been an application “output device, reading device, information processing system, output program, information processing program, computer-readable recording medium on which an output program is recorded” (see Patent Document 3). This method has a problem that the information recording efficiency is not good because only information “1” or “0” can be described using the arrangement pattern of five dots.

  In addition, an “information processing apparatus, information processing method, and information processing program” (see Patent Document 4) has been filed. By improving the above-described method, a bit “1” or a bit “0” is represented by a different arrangement direction of dots indicated by two black circles with respect to a white background area of a paper document. Bit “0” can be represented when the number of black pixels in each region is odd, and bit “1” can be represented when the number is black. In this method, when information is embedded in a white region, The change of the length of the dot pattern row according to the width of the dot pattern is not taken into consideration. In addition, there remain problems with the method of embedding a character area, such as character deterioration and inefficiency of embedding.

  Furthermore, “information embedding device, information embedding method, information embedding program and recording medium” (see Patent Document 5) was also filed, but following the above method, using the same two dot arrangement pattern, Information "1" or "0" can be described, but there is no mention of a method for specifying the position of a small dot pattern that constitutes a background pattern and a method for recognizing a dot pattern.

  In order to improve the above-mentioned problem, an “information extraction method, information extraction apparatus, information extraction program, and storage medium” (see Patent Document 6) has also been filed. Find the regression lines corresponding to the four sides of the rectangular area where the information is embedded in the dot pattern, and translate each regression line toward the inside of the rectangular area by the dot pattern size, so that the internal dot pattern Announced a method to find the regression line and determine the position of the dot pattern from the intersection of the vertical and horizontal regression lines, but when the density of the dot pattern increases, the vertical and horizontal directions may differ depending on causes such as image distortion and paper elongation. In some cases, it is difficult to accurately determine the position of the dot pattern by calculating the regression line.

  In addition, “information embedding device, information embedding method, information embedding program, information extraction device, information extraction method, information extraction program, and recording medium” (see Patent Document 7) has been filed. In this application, it is considered that information “1” or information “0” is represented by two kinds of arrangement patterns of the same two dots vertically or side by side. Is ambiguous. Here, the said minute dot was defined as a dot smaller than the limit of the image reproduction capability of the image forming apparatus. Since such fine dots may become impossible to reproduce an image, a practical problem arises. In addition, when the information is described as a collective block, a processing method for rotating the print medium is not said. Further, the result of determining the position of the dot pattern from the regression line, identifying the dot pattern, and extracting information is unlikely to be enough to cope with the problem of generating large noise when superimposed on the print data. In this method, when considering copy tolerance, the second pattern is defined using the first pattern group and the dot size expanded from the first pattern, but only the dot size is considered for the second pattern group. I didn't say about the correspondence of dot patterns in different formats.

  There has also been a recently published application “Information processing apparatus and control method thereof, image processing system, computer program, storage medium” (see Patent Document 8). This application has announced that information is described by a plurality of frequency components, but it is a method of describing information by two kinds of dot patterns of four small dots that add the same large dot and have the same gradation. Has already become a known technique (see Patent Document 9), there is a problem of novelty of this method. In addition, when embedding information in an empty area of a paper document, neither the method of changing the length of the dot pattern row according to the area size nor the configuration of the information group is discussed.

  In addition, an application “Preparation apparatus for copy-forgery-inhibited pattern pattern original, copy-forgery-inhibited pattern original, image processing apparatus and image processing method” (see Patent Document 10) has been published. In this application, information can be embedded by forming a plurality of information embedding marks having different gradations. However, as in the above-mentioned application (see Patent Document 8), the same is added to one large dot. Since the method of describing information by two kinds of dot patterns of four small dots having a smooth gradation has already become a known technique (see Patent Document 9), there is a problem of novelty of this method. In addition, when embedding information in an empty area of a paper document, a method of changing a dot pattern row depending on the area size and a configuration of an information group are not discussed. Furthermore, the efficiency of information description is poor for many irrelevant dots arranged around the information embedding mark, and the information embedding method using a plurality of background patterns having different gradations is as uniform as possible. Distant from the common sense of pursuing the composition of gradational tint blocks.

  As a completely different method from the above, “an image forming method, an information embedding method, an information embedding device, an information decoding method, and an information decoding device” (see Patent Document 11) and “a verification information embedding device, a verification information embedding program, for falsification verification” An application “Document creation device, falsification verification document creation program, falsification verification device, and falsification verification program” (see Patent Document 12) has been published. In these methods, it has been proposed to embed information by using a color material that does not absorb light in a non-visible light region, and by embedding information with a dot pattern that is subjected to density conversion and small line segments with different directions. Since ink or special toner is required, it is difficult to apply to ordinary multifunction devices.

  In all of the above-mentioned applications, the dot pattern composed of the background pattern for embedding information, the gradation characteristics of the screen halftone, the size characteristics of the screen halftone, the screen halftone spacing characteristics, etc. It is not said that the characteristics are fully considered. Further, it does not touch on important issues such as maximizing the density of embedded information and minimizing the density (tone value) of embedded dots in a background pattern.

JP-A-6-333086 Japanese Patent Laid-Open No. 2001-14422 JP 2006-237742 A JP 2006-222572 A JP 2007-258983 A JP 2006-238119 A JP 2007-282200 A JP 2007-228418 A Japanese Patent No. 4054339 JP 2008-28781 A Japanese Patent Laid-Open No. 2007-6134 JP 2007-281539 A

  The problem to be solved by the present invention is to discover a method for constructing a dot pattern and a ground pattern that embeds a large amount of information on paper, thereby maximizing the information embedding capacity and minimizing the deterioration of the image quality of the paper document. It is.

In order to solve the above problems, an invention according to claim 1 is an information embedding device having an input unit, an embedding unit, and an output unit, an information recognition device including an image input unit, a recognition unit, and a registration unit, An automatic reading system comprising: an automatic reading printed matter that prints a print image in which information is embedded;
The input unit of the information embedding device has means for inputting additional information data to be embedded and acquiring a print image as an embedded medium,
The embedding unit of the information embedding apparatus converts the input additional information data to be embedded into the screen halftone characteristics, the screen halftone size characteristics, the screen halftone spacing characteristics, and the embedded information density ratio. Means for converting into a screen code that is a pattern-recognizable geometrical or physical information embedding code that has the characteristics of a screen having at least one of the characteristics,
The configuration of the screen code arrangement includes the direction information of the dot pattern belonging to the group, the length information of the dot pattern row belonging to the group, the order information of the group, the position information of the group, the group format information, the error correction information, and the dots of different formats A pattern and a means for composing a new print image by arranging in a matrix on the print image as an information group comprising at least one type of dot patterns of different sizes;
The output unit of the information embedding device outputs the new print image,
The auto-read print is printed with a new print image output from the embedding device,
The image input unit of the information recognition device reads the automatically read printed matter as image data from the image sensor,
The recognition unit of the information recognition device has means for detecting the position of the screen code and recognizing the code value of the screen code,
The registration unit of the information recognition apparatus includes means for registering and processing the result of recognizing the screen code in at least one type of recording medium among a memory, a hard disk, a CD-ROM, and a network server. It is.

  According to a second aspect of the present invention, there is provided means capable of composing a new dot pattern by dividing at least one dot into two or more dots with respect to the dot in which the dot pattern of the screen code is configured. It is characterized by providing.

  With such a configuration, when a dot pattern of a screen code is printed using a printer with a low resolution such as 600 dpi in the example, two or more printing minimums are performed for one dot by sampling theory. Since it is necessary to configure the unit dot, it is better if the size of the information dot is small. Based on the size characteristics of the screen halftone and the gradation characteristics of the screen halftone, the dots belonging to the dot pattern are selected. By using a method of dividing a plurality of dots, a background pattern having a minimum gradation value of a dot level of a minimum printing unit can be formed. In addition, the result of dividing the dots has a feature that the embedded information can be recognized accurately because the sampling theory is not destroyed for the printed dots dispersed in a plurality of minimum printing units.

  According to a third aspect of the present invention, means for arranging the dots constituting the dot pattern of the screen code as a pattern-embedded geometrically embedded code having the characteristics of the screen, means for arranging in different directions, means for arranging in different directions It is characterized in that it is made by at least one means among means for arranging by different shapes, means for arranging by different sizes, and means for arranging by concentration and dispersion.

  By adopting such a configuration, a dot pattern of a screen code is formed as a geometrically embeddable code having a screen recognition characteristic with respect to the overall geometric dot pattern. The result of embedding the characteristic as an information embedding code by integrating the characteristic with the characteristic of the geometric dot pattern does not give a sense of incongruity to the image quality of printing.

  According to a fourth aspect of the present invention, the means for arranging the dots constituting the dot pattern of the screen code as a pattern-recognized physical embedded code having the characteristics of the screen, means for arranging by different modulation methods, and arranged by different frequency components At least one of the following means: a means for arranging with different gradation values, a means for arranging with a phase modulation method, a means for arranging with different transport directions, and a means for arranging with different mechanical characteristics. .

  By adopting such a configuration, as with the geometric dot pattern, a physical dot pattern that makes it possible to have the characteristics of the screen with respect to the overall physical dot pattern, The result of embedding an information embedding code by integrating the characteristics of the screen with the characteristics of the physical dot pattern has a characteristic that does not cause a sense of incongruity in the image quality of printing.

  The invention according to claim 5 is characterized in that the position detection of the screen code is performed by a method of performing self-organization based on a probability scale.

  With this configuration, if the arrangement range is specified for a screen code string or screen code row, each dot of the screen code is self-organized based on the arrangement probability scale of the distance from the regression line. The result is that the position of each dot in the screen code can be detected.

The invention according to claim 6 is an information embedding device having an input unit, an embedding unit, and an output unit,
The input unit of the information embedding apparatus has a module for inputting additional information data to be embedded and acquiring a print image as an embedded medium,
The embedding unit of the information embedding apparatus converts the input additional information data to be embedded into the screen halftone characteristics, the screen halftone size characteristics, the screen halftone spacing characteristics, and the embedded information density ratio. A module for converting into a screen code which is a pattern-recognizable geometrical or physical information embedding code having the characteristics of a screen having at least one characteristic among the characteristics;
The configuration of the screen code arrangement includes the direction information of the dot pattern belonging to the group, the length information of the dot pattern row belonging to the group, the order information of the group, the position information of the group, the group format information, the error correction information, and the dots of different formats A module that configures a new print image by arranging in a matrix on the print image as an information group comprising at least one type of dot patterns of different patterns and sizes;
The output unit of the information embedding device is an information embedding device that outputs the new print image.

An invention according to claim 7 is an information recognition apparatus including an image input unit, a recognition unit, and a registration unit,
The image input unit of the information recognition apparatus has a module that reads an automatically read printed material as image data from an image sensor,
The recognition unit of the information recognition device has a module that detects the position of the screen code and recognizes the code value of the screen code,
The registration unit of the information recognition apparatus includes a module for registering and processing the result of recognizing the screen code in at least one type of recording medium among a memory, a hard disk, a CD-ROM, and a network server. Device.

The invention according to claim 8 is an automatic reading printed matter that enables automatic reading,
In the automatic reading printed matter, the additional information data to be embedded is at least one of the following: gradation characteristics of screen halftone, screen halftone size characteristics, screen halftone spacing characteristics, embedded information density ratio characteristics. It is converted into a screen code that is a pattern-recognizable geometric or physical information embedding code that combines the characteristics of a screen with two characteristics.
The structure of the screen code arrangement is different in the direction information of the dot pattern belonging to the group, the length information of the dot pattern row belonging to the group, the order information of the group, the position information of the group, the group format information, the error correction information, and so on. Automatic reading characterized in that a new print image is constructed and printed by arranging in a matrix on a print image by an information group comprising at least one kind of dot patterns and dot patterns of different sizes. Printed matter.

As an invention according to claim 9, an input unit processing method for inputting additional information data to be embedded into the information embedding device according to claim 1 or 6 and acquiring a print image as an embedded medium Have
The additional information data to be embedded is converted into at least one of the following: gradation characteristics of screen halftone dots, screen halftone dot size characteristics, screen halftone dot spacing characteristics, embedded information density ratio characteristics A method of converting into a screen code that is a pattern-recognizable geometrical or physical information embedding code having the characteristics of a screen equipped with
The configuration of the screen code arrangement includes the direction information of the dot pattern belonging to the group, the length information of the dot pattern row belonging to the group, the order information of the group, the position information of the group, the group format information, the error correction information, and the dots of different formats As an information group comprising at least one type of dot patterns having different patterns and sizes, a processing method of an information embedding unit that forms a new print image by arranging the print image in a matrix, and
An information embedding program that causes a computer to execute an output unit processing method for outputting a new print image.

As an invention according to claim 10, there is provided an input unit processing method for reading an automatically read printed matter as image data from an image sensor with respect to the information recognition apparatus according to claim 1 or 7,
It has a processing method of a recognition unit that detects the position of the screen code and recognizes the code value of the screen code,
Information obtained by causing a computer to execute a registration unit processing method for registering and processing a result of recognizing a screen code in at least one type of recording medium among a memory, a hard disk, a CD-ROM, and a network server. It is a recognition program.

Here, the terms used in the present invention are defined as follows.
Pixels are the smallest unit constituting an image, halftone dots are the smallest unit constituting an image, that is, the smallest unit of a screen corresponding to a pixel, halftone dots are the smallest unit of dots constituting a halftone dot, and halftone dots are formed The printed dot is the minimum dot (1 inch / dpi) that can be printed with respect to the printing accuracy (dpi) of the printing facility.

  The minimum cell is a necessary minimum area occupied by halftone dots representing information, and the micro cell is an area corresponding to the size of the print dot.

  Geometric information embedding code means that each dot of the dot pattern with screen code is arranged by different positions, arranged by different directions, arranged by different shapes, arranged by different sizes, arranged by concentration and dispersion It is a geometric dot pattern that generically refers to these.

  The physical information embedding code is arranged with a different modulation method, arranged with a different frequency component, arranged with a different gradation value, arranged with a phase modulation method, and different for each dot of the dot pattern in which the screen code is configured. It is a physical dot pattern that collectively refers to the arrangement according to the transport direction and the arrangement according to different mechanical characteristics.

  The screen halftone gradation characteristic is the same as the screen halftone gradation uniformity, that is, the total number of screen halftone dot dots is configured in the same way, or the screen halftone gradation minimum Characteristic, that is, maximization of the ratio between the amount of embedded information and the gradation value for a fixed printing area on the paper.

The halftone dot size characteristic is a characteristic that minimizes the halftone dot size (number of halftone dots).
Here, the size of halftone dots (number of halftone dots) is set to four dots or less.

  The screen halftone dot distance characteristic means that the screen halftone dots are arranged at a fixed distance.

  The density of embedded information is the number of bits in which information is embedded in a certain area.

  The density (gradation value) of the embedded dots is the total area number of halftone dots in which information is embedded with respect to a certain area.

The ratio of the embedded information density is the number of bits in which information can be embedded in one printing dot (microcell). That is, it is a calculation result of embedded information density / embedded dot density (tone value). The larger this value, the better depending on the characteristics of the screen.
Here, the ratio of the embedded information density is set to 0.5 or more.

The human visual characteristic of the halftone dot size is based on the result calculated based on the Rayleigh criterion. The halftone dot composition is invisible with respect to the dot size and the dot interval of a dot having a diameter of 0.1 mm or less. The size.

  The arrangement reference dot is the arrangement reference information of the position of the halftone dot, the arrangement reference information of the direction, the arrangement reference information of the size, the arrangement reference information of the phase, or the conveyance direction among the dots constituting the screen halftone dot Among the arrangement reference information, one or more dots having at least one arrangement reference information having physical arrangement reference information or physical arrangement reference information are referred to as arrangement reference dots.

  A different modulation method is an amplitude modulation screen, that is, an AM screen, in which a screen halftone is represented by the size of a halftone dot with respect to a print screen halftone, and the screen halftone is represented by the number of halftone dots. Is a frequency modulation screen or FM screen. In this method, information can be described by using different modulation schemes for halftone dots of the same gradation. In addition, a dot pattern composed of different frequency components, a dot pattern composed of different dot gradation values, a dot pattern composed of different dot sizes, and a dot concentration There are various interpretation methods such as a dot pattern constituted by dispersion, but basically the same dot pattern has the same meaning.

  In the present invention, it is possible to embed a large amount of information in a paper document by using a screen code, and the gradation value of the background pattern that is the background of the paper document is low, so that the document does not feel strange. There are features.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, exemplary embodiments of an automatic reading system, an information embedding device, an information embedding program, an information recognition device, an information recognition program, and an automatic reading printed matter according to the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a block configuration example of a multi-function peripheral having an automatic reading function using a dedicated IC chip. As shown in FIG. 1, reference numeral 100 denotes a multifunction machine with an automatic reading function; 101, a scanner; 102, a printer; 103, a chip that recognizes a screen code; 104, a chip that embeds information as a screen code; A processing chip 106 of the conventional multifunction peripheral is an interface for connecting the multifunction peripheral and the computer. The scanner 101, the printer 102, the chip 103 for recognizing the screen code, the chip 104 for embedding information, the processing chip 105 of the multifunction device and the interface chip 106 may be incorporated as separate chips in the multifunction device. May I. Reference numeral 107 denotes an automatically read printed material in which information is embedded, and reference numeral 108 denotes a computer.

As shown in FIG. 1, additional information is embedded in the print data output from the computer 108 by the information embedding chip through the interface chip 106, and then output by the printer 102 to create the automatically-read printed matter 107 in which the information is embedded.
The image of the automatically-read printed matter 107 in which information is embedded is input from the scanner 101, the embedded information is recognized by the chip 103 that recognizes the screen code, and is registered in the computer through the interface chip 106 and processed.

  FIG. 2A is an example of a block configuration of a multifunction machine with an automatic reading function using driver-type software. As shown in FIG. 2A, reference numeral 200 denotes a configuration of a multi-function peripheral having an automatic reading function by driver-type software, 201 a multi-function peripheral scanner, 202 a multi-function peripheral printer, and 203 an information embedding SDK program. A printer driver and a scanner driver including an information recognition SDK program, and 204 is a computer.

As shown in FIG. 2A, additional information is embedded in the output print data from the computer 204 by the information embedding SDK program of the printer driver 203, the print data is printed in a batch, and an automatically readable paper document is created. .
A paper document that can be automatically read is input by a scanner of the multifunction device, image data is created, and the recognized additional information is registered in the computer 204 and processed by the information recognition SDK program of the driver of the scanner.

FIG. 2-2 is a block configuration example of an automatic reading system configured by internal software of a computer. As shown in FIG. 2B, reference numeral 205 denotes a general-purpose multifunction peripheral, 206 denotes an information recognition apparatus using a computer, and 207 denotes an information embedding apparatus configured using a computer.
Here, the information recognition device and the information embedding device can be configured by an electronic circuit of another dedicated IC chip other than the computer, or can be configured by a net server.

As shown in FIG. 2B, the information embedding device 207 embeds the additional information converted into the screen code in the printed image of the created document, and then outputs the information collectively by the printer of the multi-function peripheral 205 through the interface. Create a printed document with additional information embedded.
The printed matter created above reads the image of the printed matter by the scanner of the general-purpose multifunction peripheral 205, registers it in the hard disk of the information recognition device through the interface, and simultaneously recognizes the screen code embedded in the image image, and the recognition result Various processes are considered, such as registering to the hard disk, printing on a distant printer through the Internet, or registering to a database.

From now on, in order to evaluate the performance of each dot pattern, the standard dot pattern is defined as follows.
In order to recognize the dot pattern code value by sampling theory, the dot size of the associated dot pattern, the shortest distance between the dots of each dot pattern, and the shortest distance between the two dot patterns are all printed in two The diameter value of the dot (microcell).

Further, with respect to the standard dot pattern, the density of embedded information, the density of embedded dots, and the ratio of embedded information density are defined as standard dot patterns as follows.
The density of embedded information is the number of bits of information that can be embedded in a certain area. The higher the numerical value of the embedded information density, the better.

  The density of embedded dots is the number of print dots (microcells) of a standard dot pattern in which information is embedded in a certain area. Based on the gradation characteristics of the printing screen, the lower the numerical value of the density of embedded dots, the better.

  The ratio of the embedded information density is the ratio between the density of embedded information and the density of embedded dots for a certain area. Based on the gradation characteristics of the printing screen, the higher the numerical value of the ratio of embedded information density, the better.

FIG. 3 shows an example of a screen code composed of concentrated halftone dots and distributed halftone dots. As shown in FIG. 3, a concentrated halftone dot (a) in which all printed dots (microcells) are concentrated to form a single dot and a dispersed halftone dot formed by dispersing at least one halftone dot. It is an example which comprises a dot pattern using (b). The concentrated halftone dot (a) is set to the information bit value “1”, and the distributed halftone dot (b) is set to “0”. Conversely, the concentrated halftone dot (a) may be set to the information bit value “0”, and the distributed halftone dot (b) may be set to “1”.
Further, in order to form a standard dot pattern, as shown in FIG. 3, the height (H ₃₀ ) of the dot 301 at the concentrated halftone dot (a) is set to the diameter value of the two printed dots (microcells), and the concentrated dot is displayed. The width (W ₃₀ ) of the dot 301 at the point (a) is set as the diameter value of one print dot (microcell). Since the dot 302 of the dispersed halftone dot (b) works only for the effect of uniforming the gradation of the tint block and is not recognized, the height H ₃₁ and the width (W ₃₁ ) of the dot 302 of the dispersed halftone dot (b) are all equal. The diameter value of one print dot (microcell).

  As shown in FIG. 3, the concentrated halftone dot (a) can be regarded as an amplitude modulation scheme (AM) screen, and the distributed halftone dot (b) can be regarded as a frequency modulation scheme (FM) screen. That is, information can be described by dot patterns with different modulation methods.

  Further, since the concentrated halftone dot (a) can be regarded as a low frequency halftone dot and the distributed halftone dot (b) as a high frequency halftone dot, the concentrated halftone dot (a) and the distributed halftone dot (b) have different frequency components. It can be said that information is described by.

Further, since the gradation value of the dots of the concentrated halftone dot (a) is large and the gradation value of each dot of the dispersed halftone dot (b) is small, information is described by dot patterns having different dot gradation values. I can say that.
In other words, since the dot size of the concentrated halftone dot (a) is large and the size of each dot of the dispersed halftone dot (b) is small, it can be said that information is described by dot patterns having different sizes.
As described above, there are various ways of saying, but since the configuration of the dot pattern is basically the same, even if there is a way of saying other than this, if it is similar to the dot pattern shape as described above, It must belong to the range.

FIG. 4 is an example in which a tint block is formed by arranging the dot pattern as shown in FIG. 3 in a matrix.
As shown in FIG. 4, one information group is configured by an array 401 having group direction information and position information and an array 402 of data to be embedded.
Similarly, a plurality of information groups can be configured such as an array 403 having group direction information and position information and an array 404 of information data to be embedded.

  The role of the array 401 or 403 having group direction information and position information represents the direction of the data array 402 or 404 of the information to be embedded. When a paper document is read by a scanner, even if it is rotated from 0 to 360 degrees, the data array 402 or 404 of information to be embedded can be recognized based on the halftone dots of the array 401 or 403 having group direction information. It is. Also, the data of the information to be embedded is based on the array 401 or 403 having each of the halftone dots of the array 402 or 404 of the information to be embedded, which is a concentrated halftone dot, a distributed halftone dot, or group position information. The code value of each halftone dot belonging to the array 402 or 404 can be recognized.

As shown in FIG. 4, as a standard dot pattern, a halftone dot of a screen having a vertical distance (H ₄₁ ) and a horizontal distance (W ₄₁ ) between the halftone dots are equal to or larger than the diameter value of two printed dots (microcells) Depending on the gradation characteristics, the larger the vertical distance (H ₄₁ ) and horizontal distance (W ₄₁ ) between the halftone dots, the smaller the gradation value of the tint block, but the smaller the amount of embedded information.
Conversely, the smaller the vertical distance (H ₄₁ ) and the horizontal distance (W ₄₁ ) between the halftone dots, the larger the tone value of the background pattern, but the greater the amount of embedded information.
The vertical distance (H ₄₁ ) and horizontal distance (W ₄₁ ) between the halftone dots may be selected according to the actual application requirements.

The density of the embedded information of the dot pattern of the screen code shown in FIG. 3, the density of the embedded dots, and the ratio of the embedded information density are evaluated as follows.
As shown in FIG. 3, assuming that the halftone dot height (H ₃₂ ) is the diameter value of six printing dots (microcells) and the halftone dot width (W ₃₂ ) is the diameter value of three printing dots (microcells). Since the dot area (minimum cell) is 18 print dots (microcells), the dot pattern of the screen code shown in FIG. The density of the embedded information is 6, that is, 6-bit information can be embedded for each unit area.

Next, regarding the density of embedded dots in the dot pattern of the screen code shown in FIG. 3, if the total area of dots for each halftone dot is calculated as two printed dots (microcells), the density of embedded dots is 12 It is.
Using the above result, the ratio of embedded information density can be calculated as 0.5.

The smaller the number of halftone dots is, and the smaller the halftone dot size is, the smaller the halftone dot size is. We propose a screen code with 2 dots. As shown in FIG. 5, information can be embedded by using different arrangements of halftone dots 501 and 502 in four directions.
As shown in FIG. 5, if the two dots of the halftone dot (a) are arranged horizontally, the multi-bit value “0” of the information is arranged, and if the two dots of the halftone dot (b) are arranged obliquely to the left, the information If the multi-bit value “1” of the halftone dot and the two dots of the halftone dot (c) are arranged vertically, the multi-bit value “2” of the information and the two dots of the halftone dot (d) are arranged diagonally to the right. For example, the multi-bit value “3” of information can be represented. Various methods for defining multi-bit values of information are also considered, depending on the arrangement of two different dots in the four directions of halftone dots. In addition, the direction of two dots in a halftone dot is not limited to four, but may be six or eight directions.

As shown in FIG. 5, as the standard dot pattern dot with two dots, if the height H ₅₀ of each dot of the halftone dot diameter values of the two print dots (microcell), each dot of the halftone If the width W ₅₀ is the diameter value of one printing dot (microcell) or the height H _{50 of} each dot is the diameter value of one printing dot (microcell), the dot of the dot Each width W ₅₀ may be the diameter value of two printed dots (microcells). Further, the shortest distance between two dots of halftone dots is set as the diameter value of two printed dots (microcells).
The reason for this configuration is that it is necessary to recognize two dots in a halftone dot, so the size of two dots and the diameter value of two or more printed dots (microcells) are required according to sampling theory. Become.
Since each of the above values is defined as the smallest value for the standard dot pattern, it may be determined in consideration of the capacity of embedded data, the gradation value of the background pattern, and the like when actually applied.

The screen code of a halftone dot with two dots shown in FIG. 5 can be regarded as a screen code constituting a physical arrangement in different transport directions depending on the dots 501 and 502 as shown in FIG. In FIG. 6, halftone dot format a is the transport direction d ₁ , halftone dot format b is the transport direction d ₂ , halftone dot format c is the transport direction d ₃ , and halftone dot format d is the transport direction d ₄ . To do.

In this implementation method, a screen code in which the number of dots in a halftone dot is 2 as shown in FIG. 5 is different from that in FIG. It is possible to consider it as information embedding by the physical dot pattern.

Let {ζm, n} be image data that is a square standardized grid with an interval T formed by arranging screen halftone dots in a matrix, and the phase modulation formula of the screen code appears as follows.
(Equation 1)
∞ ∞
φm, n = ΣΣζ _{mi, nj} * η {[i + ε (m, n)] * T, [j + δ (m, n)] * T}
^{i =} -∞ ^{j =} -∞

Here, by changing ε (m, n) and δ (m, n), phase modulation using {ζm, n} as a carrier signal can be realized.

  As shown in FIG. 7, the dot coordinates of the halftone dot in the screen code are set to (m, n) and the computer information is set to ω (ω∈0, 1,..., K). For dot 501, halftone dot a is m = −2, n = 0; halftone dot b is m = −2, n = 2; halftone dot c is m = 0, n = 2; halftone dot d is m = 2, n = 2. For halftone dot 502, halftone dot a 'is m = 2, n = 0; halftone dot b' is m = 2, n = -2; halftone dot c 'is m = 0, n = -2; The halftone dot d ′ is m = −2 and n = −2.

  As described above, the dot pattern of FIG. 5 can be expressed in various ways, such as a dot pattern having a different dynamic vector, but it is within the scope of the present invention as long as the dot pattern configuration is the same as described above. I have to belong.

FIG. 8 shows an example in which a tint block is formed by the matrix arrangement of screen codes shown in FIG. As shown in FIG. 8, an information group is configured by an array 801 and an array 802. The embedded data code length array 801 describes the code length information of the embedded information data array 802. When embedding information in a gap between paper documents, the code lengths of the variable length codes are arranged as an array 801 in accordance with the gap size. It is also considered that error correction information is arranged in the embedded information data array 802.
Similarly, the configuration of other information groups can be configured as separate information groups as in the examples of the arrays 803 and 804 or the arrays 805 and 806.

As shown in FIG. 8, as a standard dot pattern, the halftone dot of the screen code having two dots has a vertical interval H ₈₁ = H ₈₂ and a horizontal interval W ₈₁ = W ₈₂ in the same manner. Although the diameter value is (microcell), a value larger than this may be set based on the characteristics of the printing screen in accordance with the application specifications.

The density of the embedded information of the dot pattern of the screen code shown in FIG. 5, the density of the embedded dots, and the ratio of the embedded information density are evaluated as follows.
As shown in FIG. 5, assuming that the halftone dot height (H ₅₁ ) and halftone dot width (W ₅₁ ) are the diameter values of six printed dots (microcells), the dot area (minimum cell) is Since there are 36 print dots (microcells) and the multi-bit value of the information embedding capacity of one halftone dot is 2, FIG. 5 shows 108 print dots (microcells) per unit area. The density of the embedded information of the dot pattern of the screen code is 6, that is, it is possible to embed 6-bit information for each unit area.

Next, regarding the density of embedded dots in the dot pattern of the screen code shown in FIG. 5, if the total area of dots for each halftone dot is calculated as four printed dots (microcells), the density of embedded dots is 12 It is.
Using the above result, the ratio of embedded information density can be calculated as 0.5.

  FIG. 9 shows a configuration example of a 7-base multi-bit screen code in which the dot number of halftone dots proposed by the present invention is three. As shown in FIG. 7, the dots 901 and 902 are used as arrangement reference dots, the dots 903 are used as information dots, and the cells 904 are used as cells where information dots 903 can be arranged. In FIG. 9, the dot arrangement of a halftone dot is a multibit value “0”, the dot arrangement of a b halftone dot is a multibit value “1”, and the dot arrangement of a c halftone dot is a multibit value “2”. , The dot arrangement of the halftone dot is a multibit value “3”, the dot arrangement of the halftone dot is a multibit value “4”, the dot arrangement of the f halftone dot is a multibit value “5”, The dot arrangement is a multibit value “6”.

  The screen code configuration shown in FIG. 9 is characterized in that the information dot 903 is arranged in the middle of the arrangement reference dots 901 and 902, so that misrecognition is less likely to occur due to paper stretch or misalignment. . Further, although the arrangement location of the information dot 903 varies depending on the information content, the arrangement pattern of the arrangement reference dots 901 and 902 does not depend on the code value of the screen code.

As shown in FIG. 9, as the standard dot pattern, the height H ₉₀ of the information dot 903 and the arrangement reference dots 901 and 902 is similarly set to the diameter value of two print dots (microcells), and the width W ₉₀ is set as one print. Although the diameter value of the dot (microcell) is set, it is possible to set it to a size larger than this in accordance with the specification of the application.

  FIG. 9 shows a screen code in which codes a, b, c, d, e, f and g are formed by different geometric shapes, and codes a, b, c, d, e, f and g are The screen code constituted by different geometric positions of the information dots 903, and the codes b and e, d and f can be regarded as screen codes constituted by different geometric directions.

FIG. 10 shows a dot pattern in which the screen code shown in FIG. 9 is constituted by dot transport signals of different physical phase modulation of information dots 903. As shown in FIG. 10, halftone dot (a) is m = 3, n = -4; halftone dot (b) is m = 3, n = 0; halftone dot (d) is m = -3, n = 4; The halftone dot (e) is m = -3, n = 0; the halftone dot (f) is m = -3, n = -4; and the halftone dot (g) is m = 0, n = 0.
Similarly, the arrangement reference dots 901 and 902 are dot patterns constituted by two constant phase dot carrier signals, and the codes a, b, c, d, e, f and g shown in FIG. Can also be regarded as a physical dot pattern in which two constant phase dot carrier signals and phase modulated dot carrier signals are superimposed.

  In contrast to the screen code having the geometric or physical dot pattern arrangement characteristics shown in FIG. 9 together with the screen characteristics, the screen characteristics of the halftone dots constituting the screen code are different even if the code values are different. The gradation is uniform, and the smaller the number of dots or the smaller the dot size, the less influence the result of embedding in the printed image has on the printed image. .

  FIG. 11 is an example of the arrangement of halftone dots that form a background pattern by arranging the screen codes of the three dots of FIG. 9 in a matrix based on the characteristics of the screen. As shown in FIG. 11, halftone dots (a) to (g) are information-embedded data, and halftone dots (h) and (i) are halftone dots of a screen code different from the normal dot size. When embedding additional information such as copy prohibition or tracking in confidential materials, the size of the normal halftone dot is used to prevent illegal persons from using the reduced copy function. Since it is necessary to use a dot pattern composed of large dots, it is a countermeasure method to mix dots of different sizes in one information group.

As shown in FIG. 11, as a standard dot pattern, the vertical interval (H ₁₁₁ = H ₁₁₂ ) and the horizontal interval (W ₁₁₁ = W ₁₁₂ ) between two halftone dots from halftone dots (a) to (i) are each zero. The diameter value of the print dot (microcell) is used, but a value larger than that may be set according to the application specification.
Further, a vertical interval (H ₁₁₁ ≠ H ₁₁₂ ) or a horizontal interval (W ₁₁₁ ≠ W ₁₁₂ ) between the halftone dots may be set, and the vertical or horizontal interval between the halftone dots of the information group may be set based on the print screen characteristics. On the other hand, it should be set to a constant value.

  As described above, for the background pattern composed of screen codes, if the distance between the halftone dots is wide for the same gradation, the overall gradation of the background pattern is small. The effect on the image quality of the printed image is reduced.

The density of embedded information, the density of embedded dots, and the ratio of embedded information density in the dot pattern of the screen code shown in FIG. 9 are evaluated as follows.
As shown in FIG. 9, the dot height (H ₉₁ ) is the diameter value of 12 printing dots (microcells), and the dot width (W ₉₁ ) is the diameter value of nine printing dots (microcells). Then, the area of the halftone dot (minimum cell) is 108 print dots (microcell), and the code length of the screen code is 7. Therefore, with respect to 108 print dots (microcell) of unit area, The density of the embedded information of the dot pattern of the screen code shown in FIG. 9 is about 3, that is, about 3 bits of information can be embedded for each unit area.

Next, regarding the density of embedded dots in the dot pattern of the screen code shown in FIG. 9, if the total area of dots for each halftone dot is calculated as six printed dots (microcells), the density of embedded dots is 6 It is.
Using the above result, the ratio of embedded information density can be calculated as about 0.5.

  In the present invention, in order to increase the information amount of the screen code as shown in FIG. 9, a 20-bit multi-bit screen code in which the number of dots of a halftone dot is 4 is proposed as shown in FIG. As shown in FIG. 12, the dots 1201 and 1202 are used as arrangement reference dots, and the dots 1203 and 1204 are used as information dots. In FIG. 12, the dot arrangement of the halftone dot is a multibit value “0”, the dot arrangement of the halftone dot is a multibit value “1”, and the dot arrangement of the halftone dot is a multibit value “2”. , The dot arrangement of the halftone dot is a multibit value “3”, the dot arrangement of the halftone dot is a multibit value “4”, the dot arrangement of the f halftone dot is a multibit value “5”, The dot arrangement is a multibit value “6”, the dot arrangement of the h halftone dot is a multibit value “7”, the dot arrangement of the i halftone dot is a multibit value “8”, and the dot arrangement of the j halftone dot is multibit. The value “9”, the dot arrangement of the k halftone dot is the multibit value “A”, the arrangement of the halftone dot is the multibit value “B”, the dot arrangement of the m halftone dot is the multibit value “C”, The dot arrangement of n halftone dots is the multibit value “D”, and the dot of o halftone dot The multi-point value is “E”, the dot arrangement of the p halftone dot is the multi-bit value “F”, the dot arrangement of the q half-tone dot is the multi-bit value “G”, and the dot arrangement of the r half-tone dot is the multi-bit value. It is assumed that “H”, the dot arrangement of the s halftone dot is the multibit value “I”, and the dot arrangement of the t halftone dot is the multibit value “J”.

  Similarly to the three-dot screen code shown in FIG. 9, the halftone dots a, b, c, d, e, f, g, h, i, j, k, l, m, n, and the like shown in FIG. o, p, q, r, s and t are screen codes configured by different geometric shapes, or codes a, b, c, d, e, f, g, h, i, j, k. , L, m, n, o, p, q, r, s and t are screen codes formed by different geometric positions of information dots 1203 and 1204, or halftone dots a and p, g and r. , Halftone dots c and n, halftone dots d and h, j and m, halftone dots e and l, halftone dots f and q, o and t, halftone dots k and s are composed of different geometrical directions. It can be regarded as a code.

  Similarly to the three-dot screen code shown in FIG. 9, the screen codes a, b, c, d, e, f, g, h, i, j, k, l, m, as shown in FIG. n, o, p, q, r, s, and t are two constants, that is, a dot pattern constituted by dot transport signals of different physical phase modulation of information dots 1203 and 1204 and dots 1201 and 1202 of arrangement reference. It can also be regarded as a physical screen code on which a dot pattern constituted by a phase dot carrier signal is superimposed.

  Further, like the three-dot screen code shown in FIG. 9, the screen code shown in FIG. 12 is a configuration of a pattern-recognizable geometric or physical information embedding code having characteristics of the screen. In other words, it is an information embedding code in which a screen characteristic is added to the configuration of a geometric or physical code. As for the characteristics of the screen of halftone dots constituting the screen code, the gradations of the halftone dots are all uniform even with different code values. In addition, the amount of information recorded in the screen code is secured, and the smaller the number of information dots and the smaller the halftone dot size, the less the embedding result has an effect on the printed image. Was used.

FIG. 13 shows an example of a background pattern configured by a matrix arrangement of four-dot screen codes.
As shown in FIG. 13, one information group is composed of a screen code including a group order information bit group 1301, an embedded information group 1302, an error correction information group 1303, and the like. In order to represent the arrangement direction of the group, one screen code example 1304 is rotated 90 degrees for every four neighboring screen codes. The merit of this method can be read even if the document is rotated 360 degrees arbitrarily.

The density of embedded information, the density of embedded dots, and the ratio of embedded information density in the dot pattern of the screen code shown in FIG. 12 are evaluated as follows.
As shown in FIG. 12, the halftone dot height (H ₁₂₁ ) is the diameter value of 12 printing dots (microcells), and the halftone dot width (W ₁₂₁ ) is the diameter value of nine printing dots (microcells). Then, the area of the halftone dot (minimum cell) is 108 printed dots (microcells), and the multi-bit value of the information embedding capacity of one halftone dot is 4.25. The density of the embedding information of the dot pattern of the screen code shown in FIG. 12 is 4.25, that is, it is possible to embed information of 4.25 bits per unit area with respect to the printing dot (microcell).

Next, regarding the density of embedded dots in the dot pattern of the screen code shown in FIG. 12, if the total area of dots for each halftone dot is calculated as eight printed dots (microcells), the density of embedded dots is 8 It is.
Using the above result, the ratio of the embedded information density can be calculated as 0.53. The ratio of the embedded information density of the dot pattern of the screen code shown in FIG. 12 is higher than that of the screen code shown in FIGS. 3, 5, and 9. This effect takes into account both the print image quality and the capacity of the embedded information. This is the result obtained.

As another form of the method for constructing the screen code shown in FIG. 9 or FIG. 12, in the present invention, the smaller the number of dots and the smaller the dot size, the more the embedded result is a printed image. On the other hand, based on the characteristics of the printing screen having little influence, as shown in FIG. 14, the dots of the arrangement reference as shown in FIG. 9 or FIG. In addition, a method of constructing a multi-bit screen code in decimal notation in which the dot number of halftone dots is 1 is proposed.
As shown in FIG. 14, 1401 is an information dot, and 1402 is a cell in which the information dot 1401 can be placed. First, it is considered that the information dot 1401 describes information by arrangement of geometrically different positions. In the example shown in FIG. 14, the dot arrangement of the halftone dot is set to the multibit value “0”, the dot arrangement of the halftone dot is set to the multibit value “1”, and the dot arrangement of the halftone dot is set to the multibit value “2”. , The dot arrangement of the halftone dot is a multibit value “3”, the dot arrangement of the halftone dot is a multibit value “4”, the dot arrangement of the f halftone dot is a multibit value “5”, The dot arrangement of the dots is a multibit value “6”, the dot arrangement of the h halftone dot is a multibit value “7”, and the dot arrangement of the i halftone dot is a multibit value “8”.
Referring to the example of FIG. 14, a multi-bit screen code having a halftone dot number of 1 or more and a 32-bit system or more can be configured by arranging different positions of information dots. .

  Similarly to the screen codes shown in FIGS. 9 and 12, screen codes a, b, c, d, e, f, g, h, and i as shown in FIG. 14 are dot carrier signals of different phase modulation of information dots 1401. Is regarded as a physical dot pattern constituted by It can also be regarded as a physical dot pattern composed of different dynamic centers of gravity.

In FIG. 15, the 9-bit multi-bit screen code in which the number of dots of halftone dots shown in FIG. 14 is 1, the background pattern is formed by arranging a plurality of screen codes as information groups in a matrix based on the characteristics of the screen. FIG. 1501 is an information dot, and 1502 is a cell in which the information dot 1501 can be placed.
As shown in FIG. 15, a, b, c, d, e, f, g, h, i, j, k, l, m, n are arranged for the screen code arrangement method considering the screen characteristics. , O and p 16 screen code halftone dots are formed as one information group. As a standard dot pattern, the information dot 1501 has a height (H ₁₅₁ ) as the diameter value of two printing dots (microcells) and a width (W ₁₅₁ ) as the diameter value of one printing dot (microcell). . The dot size of a screen code of one dot is such that the height (H ₁₅₂ ) is the diameter value of 12 printing dots (microcells) and the width (W ₁₅₁ ) is nine printing dots (microcells). The diameter value is set, and the distance (H ₁₅₃ ) and (W ₁₅₃ ) between the two halftone dots are also set as the diameter value of the zero print dot (microcell).
Depending on the actual application specifications, the size may be selected larger than the above based on the screen characteristics.

Reference numeral 1502 in FIG. 15 denotes a cell in which information dots can be arranged in a halftone dot of a screen code of one dot. As shown in FIG. 15, the halftone dots a, b, c, and d are all arranged as information reference dots in the middle. The direction of the information group can be recognized using the halftone dots of a, b, c and d, which are halftone dots representing the halftone dots of a, b, c and d in the direction of the information group. At the same time, the halftone dots a, b, c, and d are halftone dots having position reference information of information dots in the vertical direction.
Further, as shown in FIG. 15, e, i, and m halftone dots are halftone dots having position reference information of information dots in the horizontal direction. For e, i, and m halftone dots, the halftone dots of e, i, and m are all empty in the middle, and information dots can be arranged on either side of the grid.
Based on the positions of the halftone dots a, b, c and d having the reference information of the vertical information dots and the positions of the halftone dots e, i and m having the position reference information of the horizontal information dots, the halftone dots f, Multicode values of g, h, j, k, l, n, o and p can be recognized.
Further, in order to distinguish from the arrangement of halftone dots e, i, and m having the position reference information of the information dots in the horizontal direction, in the middle of the halftone dots f, j and n and g, k and o and h, l and p. Do not leave blank at the same time.

The density of embedding information, the density of embedding dots, and the ratio of the embedding information density of the dot pattern of one dot screen code shown in FIG. 14 are evaluated as follows.
As shown in FIG. 14, the dot height (H ₁₄₂ ) is the diameter value of 12 print dots (microcells), and the dot width (W ₁₄₂ ) is the diameter value of nine print dots (microcells). Then, the dot area (minimum cell) is 108 printed dots (micro cells), and the multi-bit value of the information embedding capacity of one dot is 3.25. The density of the embedding information of the dot pattern of the screen code shown in FIG. 14 is 3.25, that is, it is possible to embed information of 3.25 bits per unit area with respect to the printing dot (microcell).

Next, regarding the density of embedded dots in the dot pattern of the screen code shown in FIG. 14, if the total area of dots for each halftone dot is calculated as two printed dots (microcells), the density of embedded dots is 2 It is.
Using the above result, the ratio of the embedded information density can be calculated as 1.625. The ratio of the embedded information density of the dot pattern of the screen code shown in FIG. 14 is higher than that of the screen code shown in FIG. 3, FIG. 5, FIG. 9, and FIG. This is a result obtained by considering both. Here, it can be said that the result is obtained in consideration of the characteristics of the printing screen.

FIG. 16 is an example of dividing information dots belonging to a dot pattern constituting a screen code. In consideration of the printing screen characteristics, in order to further reduce the tone value of the tint block, the information dot 1600 of the dot pattern of the screen code of one dot is divided into two small information dots 1601 as shown in FIG. It is possible. Even if the size of the information dot 1601 is the diameter value of one print dot (microcell), the information dot 1601 is two dispersed information dots 1601, so that information recognition can be performed by sampling theory. A cell 1602 is an area where information dots 1601 can be arranged. Further, H ₁₆₃ is set as an interval between two dots of the information dot 1601.
Similarly, the dot pattern of the proposed screen code can be used to divide dots belonging to the dot pattern for all the dot patterns shown above with reference to FIG. it can.

With respect to the dot pattern of the screen code of the dot division shown in FIG. 16, the density of the embedded information, the density of the embedded dots, and the ratio of the embedded information density are evaluated as follows.
As shown in FIG. 16, the halftone dot height (H ₁₆₂ ) is the diameter value of 15 printing dots (microcells), and the halftone dot width (W ₁₆₂ ) is the diameter value of nine printing dots (microcells). Then, the halftone dot area (minimum cell) is 135 printed dots (microcells), and the multi-bit value of the information embedding capacity of one halftone dot is 2.6. The density of embedding information of the dot pattern of the screen code shown in FIG. 14 is 2.6 with respect to the print dots (microcells), that is, 2.6 bits of information can be embedded per unit area.

Next, regarding the density of embedded dots in the dot pattern of the screen code shown in FIG. 16, if the total area of dots for each halftone dot is calculated as two printed dots (microcells), the density of embedded dots is 1. .6.
Using the above result, the ratio of the embedded information density can be calculated as 1.625.
Here, as described above, the density of embedding information, the density of embedding dots, and the embedding information density of the dot pattern of the screen code shown in FIGS. 3, 5, 9, 12, 14, and 16 are proposed. The ratios are compared as shown in Table 1.

Since the printing screen characteristics are considered for the dot patterns of the six types of screen codes shown in FIGS. 3, 5, 9, 12, 14 and 16 proposed above, the evaluation of the information density ratio value is performed. As a result, an excellent result that is close to 0.5 or more than 0.5 was obtained.
In other words, one important characteristic among the characteristics of the printing screen is that the ratio of the information density of the standard dot pattern of the dot pattern is 0.5 or more.

  Further, with respect to the dot patterns of the six types of screen codes shown in FIGS. 3, 5, 9, 12, 14, and 16 proposed above, the dot pattern configuration other than those described above is geometrical or physical. Although there is a typical interpretation method, it is similar to the dot patterns of the six types of screen codes proposed above in FIGS. 3, 5, 9, 12, 14, and 16, and further adds the characteristics of the printing screen. If it does, it must belong within the claim of this invention.

Hereinafter, the implementation method of the information embedding apparatus and the information recognition apparatus related to the automatic reading system of the present invention shown in FIG. 1 or 2 will be described in more detail.
FIG. 17 is a processing flow of the information embedding device and the information recognition device. As shown in FIG. 17, an information embedding device having an input unit, an embedding unit, and an output unit, an information recognition device having an image input unit, a recognition unit, and a registration unit, and print image data in which information is embedded are printed. An automatic reading system consisting of the automatic reading printed matter can be constructed as follows.

The input unit of the information embedding device performs the processing contents of S ₁ and S ₂ . The additional information data to be automatically read is input, the process proceeds to the screen conversion step, the document data is input, and the process proceeds to the image conversion step.

The embedding unit of the information embedding apparatus performs the processing contents of S ₃ , S ₄ and S ₅ .
The additional information data S ₃ steps to be automatically read subjects the input, the tone characteristic of the halftone screen, the size characteristics of the halftone screen, spacing characteristic of the halftone screen, the characteristics of the ratio of the embedded information density Among these, the screen code is converted into a pattern-recognizable geometrical or physical information embedding code having the characteristics of a screen having at least one characteristic. Complete the conversion screen code, the process proceeds to step S _5.

S ₄ step converts the data of the statements in the previous step input to the image data, good to convert the data of the document directly to the image data, good to retrieve the image data of the document from the printer driver. After obtaining the image data of the document, the process proceeds to step S _5.

S ₅ step performs information embedding processing, the pre-screen code converted in step as the information group, there is a method of configuring a print image of the new document by matrix arrangement in the gaps of the image data of the document.
Another method of information embedding is to change the screen code color to white for the area where the black part of the document data overlaps the screen code when the document image data is black and white data. . Also, when the document image data is color data, the screen code color is different from the color of the document data near the screen code for the area where the colored part of the document data and the screen code overlap. Change to color.

  The above information group is the direction information of the dot pattern belonging to the group, the length information of the dot pattern row belonging to the group, the order information of the group, the position information of the group, the format information of the group, the error correction information, the format within the group Compared to different dot patterns and groups, as well as halftone dots of screen codes, at least one kind of dot patterns having different sizes is provided.

The output unit of the information embedding apparatus performs the processing of documents printed Step S _6. A print image of the new document composed in the previous step is output, and an automatically read printed matter in which information is embedded is created.
Here, the automatically read printed material in which information is embedded is made by a paper slip or paper document.

Next, the information recognition apparatus includes an image input unit, a recognition unit, and a registration unit. The flow of the process has four steps of S ′ ₁ , S ′ ₂ , S ′ ₃ and S ′ ₄ .
The image input unit of the information recognition apparatus reads the automatically read printed matter from the image sensor by S ′ ₁ and sends it to step S ′ ₂ as image data.
The recognition unit of the information recognition apparatus is configured by steps S ′ ₂ and S ′ ₃ .
Step S ′ ₂ uses a self-organization method based on a probability scale to detect the screen code halftone dot column or dot row by utilizing the feature that the dot arrangement density of the dot pattern is higher than the surrounding pixels. It is to detect the position of each dot of the screen code.
Step S ′ ₃ is to recognize a multi-bit value for each dot pattern of the screen code detected in the previous step.

The registration unit of the information recognition apparatus is configured by step S ′ ₄ .
Whether the result of recognizing the screen code is registered in at least one type of recording medium in memory, hard disk, CD-ROM, or network server, printed on a remote printer over the Internet, or registered in a database Think about various processes.

FIG. 18 is a specific processing flow of the self-organization method based on the probability scale.
As shown in FIG. 18, the initial setting step S ″ ₁ initializes a detection area called a screen code halftone dot column or dot row existence area to be detected, and sets the detection area length (L _s ) to The center line is calculated as a reference line (S _L ).
Here, regarding the setting method of the rectangular detection area, the vertical or horizontal length (L _s ) of the detection area is the length of the screen code halftone dot vertical column or dot horizontal row, and the width or height (P _s ) of the detection region is set. This is a rectangular area in which the width of the screen code halftone dot column to be detected or the height of the dot row is five times or more, that is, the screen code halftone dot column or dot row to be detected is sufficiently included.

In the probability scale calculation step S ″ ₂ , the new width of the detection area or the detection area is measured with a certain probability (0.7 or more) as a scale with respect to the arrangement distance of all dots in the detection area with the reference line (S _L ). Find the height (P _s ').
In other words, the detection area is an area that can include 70% or more halftone dots.

A new dot selection step S " ₃ selects all dots belonging to the new detection area.

A new reference line (S _L ) calculation step S ″ ₄ calculates the regression line as a reference line (S _L ) for all the dots selected in step S ″ ₃ .

The determination step S ″ _{5 for the} completion of the self-organization processing is whether the new width or height (P _s ′) of the detection region and the value of the reference line (S _L ) are stable. If the process is completed (Y), the process proceeds to the next end step, and if the self-organization process is not completed (N), the process returns to the probability scale calculation step S " ₂ . If the position of the reference line (S _L ) is obtained by repeating the self-organization processing as described above, the position of the screen code halftone dot column or dot row can be detected.

  Here, a self-organization method based on a probability scale has been proposed, but there are various methods for detecting the position of a dot pattern with reference to the above-described method. Therefore, if it is similar to the above-described method, it must be considered as within the scope of claims of the present application.

In the present invention, the following business model is proposed using an automatic reading system in which a screen code is introduced.
FIG. 19 is a configuration example of a total slip automatic reading system for business forms.
Here, it is considered that the information embedding device and the information recognition device shown in FIG. 17 are configured as a network server. Further, if the additional information of the information embedding device is a document data file, the document data is embedded as a file in the image of the automatically read printed matter of the document, so that the automatically recognizable printed matter is read through the information recognition device to read the document data. Therefore, as shown in FIG. 19, a total slip automatic reading system for business forms can be configured.

  As shown in FIG. 19, an example is a scanner of an American multifunction device 1902, which reads a paper document embedded in a screen code as image data into a server 1901 and recognizes an electronic file embedded in a paper document by the server 1901. Whether to read and register in the server 1901, or to send to the Chinese server 1903 via the Internet and register in the Chinese server 1903, or to print by the multifunction device 1904. Its merit is that it can embed information of about tens of pages in an image of a single paper document, so when sending it as an electronic file, the speed is fast, and by reducing the number of paper documents by the US side, the society as a whole There are also ecological effects to be addressed.

  As shown in FIG. 19, in another example, a paper document in which information is embedded is read into a server 1903 as image data by a scanner of a Chinese multifunction device 1904, and the server 1903 transmits the image data to a Japanese server via the Internet. The screen code embedded in the image of the paper document is recognized by the server 1905 in Japan, and the electronic file of the original document is taken out and registered in the server 1905 in Japan, or the multifunction device 1906 on the Japanese side. Output from? The advantage of this method is that when sending a paper document with embedded information as it is, it is a dot pattern with a multi-adic multi-bit screen code that describes the information, so if the document data is encrypted It can be proved that the complexity of encryption can be increased exponentially, and if the document data is encrypted and further concealed, it can be said to be a very secure transmission method.

FIG. 20 is an example of a paper document tracking system. The name of documents the additional information in step S ₂ in FIG. 17, if such tracking information printing equipment number and the printing time can be as shown in FIG. 20, constituting the tracking system of the document paper.
As shown in FIG. 20, a paper document in which tracking information is embedded is printed by a printer or a printing machine 2001, and a paper document is generated. At any stage of circulation of paper documents, tracking information can be registered in the computer 2004 by using a screen code reader 2002. Finally, when a paper document is discarded, an automatic screen code reader is installed. By using the shredder 2003, it becomes possible to register the tracking information of the discarded paper document in the computer 2004.
Compared with ordinary non-contact IC tag technology, it is possible to smoothly manage a series of paper documents embedded with tracking information as screen codes at low cost.

The configuration using the above information embedding device, to create a paper document, if step the additional information printing time such as time stamp information S ₂ in FIG. 17, a time stamp authentication system paper documents can do.

Further, by using the information embedding apparatus, to create a paper certificate, if the additional information in step S ₂ in FIG. 17 and certified contents such as certificates alteration prevention information in the certificate, the certificate falsification A prevention system can be configured.

Or, using the above-described information embedding apparatus, when creating a secret document, if the additional information in step S ₂ in FIG. 17 and information such as the secret contents of confidential documents, that constitute the production system confidential documents it can.

  Finally, FIG. 21 is a sample of an example of an automatic reading printed matter using the above information embedding device. As shown in FIG. 21, the image quality of the paper document in which the additional information is embedded is almost the same as that of the conventional document, and the data file of the paper document can be embedded in the paper document as it is.

FIG. 3 is a diagram illustrating a configuration example of a multifunction peripheral having an automatic reading function using a dedicated IC chip. FIG. 2 is a diagram illustrating two examples of a configuration of a multifunction machine with an automatic reading function by software. It is a figure which shows the example of the screen code comprised by the concentration halftone dot and the dispersion | distribution halftone dot. It is a figure which shows an example which comprises a tint block by arrangement | positioning of a matrix. It is a figure which shows the example of the screen code which makes the dot number of a halftone dot two. It is a figure which shows the example of the screen code comprised by the physical arrangement | positioning of a different conveyance direction. It is a figure which shows the example of the screen code comprised by the physical arrangement | positioning of a different phase modulation. It is a figure which shows another example of the structure of the background pattern arranged in matrix. It is a figure which shows the structural example of the screen code which makes the dot number of a halftone dot three. It is a figure which shows the dot pattern comprised by different physical phase modulation. It is a figure which shows the example of the matrix arrangement | positioning of the screen code of the dot number 3 of a halftone dot. It is a figure which shows the structural example of the screen code which sets the dot number of a halftone dot to four. It is a figure which shows the example of the matrix arrangement | positioning of the screen code of the dot number 4 of a halftone dot. It is a figure which shows the structural example of the screen code which makes the dot number of a halftone dot one. It is a figure which shows the example of the matrix arrangement | sequence of the screen code of the dot number 1 of a halftone dot. It is a figure which shows the example which enables the division | segmentation of the dot of a screen code. It is a figure which shows the processing flow of an information embedding apparatus and an information recognition apparatus. It is a figure which shows the specific processing flow of the self-organization method based on a probability scale. It is a figure which shows the example of the total total reading system for business forms. It is a figure which shows the example of the tracking system of a paper document. It is a figure which shows the example of the sample of an automatic reading printed matter.

Claims (10)

An information embedding device having an input unit, an embedding unit, and an output unit; an information recognition device including an image input unit, a recognition unit, and a registration unit; an automatic reading printed matter that prints a print image in which additional information is embedded; An automatic reading system comprising:
The input unit of the information embedding device has means for inputting additional information data to be embedded and acquiring a print image as an embedded medium,
The embedding unit of the information embedding apparatus converts the input additional information data to be embedded into the screen halftone characteristics, the screen halftone size characteristics, the screen halftone spacing characteristics, and the embedded information density ratio. Means for converting into a screen code that is a pattern-recognizable geometrical or physical information embedding code that has the characteristics of a screen having at least one of the characteristics,
The configuration of the screen code arrangement includes the direction information of the dot pattern belonging to the group, the length information of the dot pattern row belonging to the group, the order information of the group, the position information of the group, the group format information, the error correction information, and the dots of different formats A pattern and a means for composing a new print image by arranging in a matrix on the print image as an information group comprising at least one type of dot patterns of different sizes;
The output unit of the information embedding device outputs the new print image,
A new print image output from the embedding device is printed on the automatic reading printed matter,
The image input unit of the information recognition device reads the automatically read printed matter as image data from the image sensor,
The recognition unit of the information recognition device has means for detecting the position of the screen code and recognizing the code value of the screen code,
The registration unit of the information recognition apparatus includes means for registering and processing the result of recognizing the screen code in at least one type of recording medium among a memory, a hard disk, a CD-ROM, and a network server. .   The dot comprising the dot pattern of the screen code is provided with means capable of forming a new dot pattern by dividing at least one dot into two or more dots. 2. The automatic reading system according to 1.   A means for arranging the dot pattern of the screen code as a pattern-embedded geometric code that has the characteristics of the screen, arranged at different positions, arranged at different directions, arranged at different shapes 2. The automatic reading system according to claim 1, wherein the automatic reading system is formed by at least one of a means for arranging with different sizes and a means for arranging with concentration and dispersion.   A means to arrange with different modulation methods, a means to arrange with different frequency components, and different gradation values for the dots with the screen code dot pattern that is a pattern-embedded physical embedding code that has the characteristics of the screen 2. The automatic operation according to claim 1, wherein at least one of the means for arranging according to the phase modulation method, the means for arranging according to different transport directions, and the means for arranging according to different mechanical characteristics is used. Reading system.   2. The automatic reading system according to claim 1, wherein the position detection of the screen code is performed by a method of performing self-organization based on a probability scale. An information embedding device comprising an input unit, an embedding unit, and an output unit,
The input unit of the information embedding apparatus has a module for inputting additional information data to be embedded and acquiring a print image as an embedded medium,
The embedding unit of the information embedding apparatus converts the input additional information data to be embedded into the screen halftone characteristics, the screen halftone size characteristics, the screen halftone spacing characteristics, and the embedded information density ratio. A module for converting into a screen code which is a pattern-recognizable geometrical or physical information embedding code having the characteristics of a screen having at least one characteristic among the characteristics;
The configuration of the screen code arrangement includes the direction information of the dot pattern belonging to the group, the length information of the dot pattern row belonging to the group, the order information of the group, the position information of the group, the group format information, the error correction information, and the dots of different formats A module that configures a new print image by arranging in a matrix on the print image as an information group comprising at least one type of dot patterns of different patterns and sizes;
The information embedding device, wherein the output unit of the information embedding device outputs the new print image. An information recognition apparatus comprising an image input unit, a recognition unit, and a registration unit,
The image input unit of the information recognition apparatus has a module that reads an automatically read printed material as image data from an image sensor,
The recognition unit of the information recognition device has a module that detects the position of the screen code and recognizes the code value of the screen code,
The registration unit of the information recognition apparatus includes a module for registering and processing the result of recognizing the screen code in at least one type of recording medium among a memory, a hard disk, a CD-ROM, and a network server. apparatus. An automatic reading printed matter that enables automatic reading,
In the automatic reading printed matter, the additional information data to be embedded is at least one of the following: gradation characteristics of screen halftone, screen halftone size characteristics, screen halftone spacing characteristics, embedded information density ratio characteristics. It is converted into a screen code that is a pattern-recognizable geometric or physical information embedding code that combines the characteristics of a screen with two characteristics.
The structure of the screen code arrangement is different in the direction information of the dot pattern belonging to the group, the length information of the dot pattern row belonging to the group, the order information of the group, the position information of the group, the group format information, the error correction information, and so on. Automatic reading, characterized in that a new print image is constructed and printed by arranging in a matrix on the printed image with at least one type of dot pattern and dot patterns of different sizes Printed matter. The information embedding device according to claim 1 or 6 includes an input unit processing method for inputting additional information data to be embedded and acquiring a print image as an embedded medium,
The additional information data to be embedded is converted into at least one of the following: gradation characteristics of screen halftone dots, screen halftone dot size characteristics, screen halftone dot spacing characteristics, embedded information density ratio characteristics A method of converting into a screen code that is a pattern-recognizable geometrical or physical information embedding code having the characteristics of a screen equipped with
The configuration of the screen code arrangement includes the direction information of the dot pattern belonging to the group, the length information of the dot pattern row belonging to the group, the order information of the group, the position information of the group, the group format information, the error correction information, and the dots of different formats As an information group comprising at least one type of dot patterns having different patterns and sizes, a processing method of an information embedding unit that forms a new print image by arranging the print image in a matrix, and
An information embedding program for causing a computer to execute an output unit processing method for outputting a new print image. The information recognition apparatus according to claim 1 or 7, further comprising an input unit processing method for reading an automatically read printed matter as image data from an image sensor,
A processing method of a recognition unit that detects the position of the screen code and recognizes the code value of the screen code;
Information obtained by causing a computer to execute a registration unit processing method for registering and processing a result of recognizing a screen code in at least one type of recording medium among a memory, a hard disk, a CD-ROM, and a network server. Recognition program.