KR20030062463A - Method for embedding and extracting watermark into/from a text document, and the apparatus thereof - Google Patents

Abstract

PURPOSE: A method and a device for embedding and extracting a watermark to and from a text file are provided to embed watermark to sentences or words in a text file without any changes distinguished by a naked eye and extract the embedded watermark in spite of clustering errors. CONSTITUTION: A device for embedding a watermark to a text file includes a sentence analyzing part(120), a region selecting part(130) for selecting same characters having different character code values in a text file, a watermark generating part(140) for generating binary watermark bit stream signals, an error correcting code(150), an extension part(160) for extending the watermark bit stream signals with R-multiples, a key inserting part(170), a random number generating part(180) for generating binary pseudo random number signals of wide band from a predetermined user key, an XOR converting part(190) for XOR-operating the extended watermark bit stream signals and the binary pseudo random number signals to generate watermark bit stream signals of diffusion spectrum type, a shuffling part(200) and a combination part(210) for embedding the watermark to the text file by the equal characters having different character code values.

Description

METHOD FOR EMBEDDING AND EXTRACTING WATERMARK INTO / FROM A TEXT DOCUMENT, AND THE APPARATUS THEREOF}

The present invention relates to a method and apparatus for embedding and extracting watermarks into / from text documents. More specifically, the present invention provides a method of inserting and extracting a watermark into and out of a text document without damaging the original of the text document by inserting and extracting the watermark using characters having the same characters but different code values, and Relates to a device.

Digital watermarking inserts watermarks, which are specific data intended to be claimed by a user's information or copyright information, into digital multimedia content such as images, audio, video, or text documents so that the human eye or hearing cannot recognize them. Says technology.

When the content distribution process needs to identify the original owner of the content in relation to ownership or copyright, a watermark embedded in the content may be extracted to provide a basis for exercising the right of ownership or copyright.

Among these, text watermarking includes inserting a watermark into a text portion of a document. For example, a method of converting text into an image file and watermarking it or changing a space between letters finely.

The method of converting text into an image file and watermarking was filed by the same applicant as the applicant of the present application, and is disclosed in Korean Patent Application No. 1999-58131 entitled "Text Watermark Insertion Method and Apparatus". . The reason that text is converted to an image file and watermarked is that the text is vulnerable to deformation or editing, and the inserted watermark may be broken.

Watermarking by converting the space between the letters is the "Open Space" method proposed by W. Bender of IBM, etc. in 1996, this will be described with reference to FIG. 1 is a diagram illustrating a conventional watermark embedding example in which a watermark is inserted into a text document by changing the space between letters.

Referring to FIG. 1, first, the space between words is adjusted so that the space of one letter is defined as '0' and the space of two letters is defined as '1'. In addition, if the space on the left and right of the word is 01 (that is, if the space on the left is one letter and the space on the right is two letters), it is interpreted as binary '0', and if the space on the left and right of the word is 10, Interpret as binary '1'. That is, the binary bit string of the watermark information inserted in the lower paragraph watermarked by adjusting the space spacing between words in FIG. 1 becomes '01011011'.

However, according to the method of watermarking by converting the spaces between the letters as described above, there is a disadvantage in that the space in the sentence is increased, and thus the total size of the text is increased and the file size is increased. In addition, the method has a disadvantage in that the changed text can be visually distinguished, thereby failing to achieve the basic purpose of watermarking of inserting information that cannot be visually identified.

Other text watermarking methods have been proposed, such as the method of adjusting the font size of letters, but all of them have the disadvantage of increasing the capacity of the text or distinguishing them visually. Since the back is easily changed when editing the text, the inserted watermark is not strong.

That is, the insertion of watermarks should not be visually identified as a watermark inserted by default. In the case of the conventional text watermarking, there is a disadvantage in that the original of the text is damaged, and the inserted watermark is easily deformed during editing. There are disadvantages.

SUMMARY OF THE INVENTION The present invention has been made to solve the problems of the prior art as described above, and an object of the present invention is to provide a method for embedding a watermark without making visually identifiable changes to sentences or words in a text document.

It is another object of the present invention to provide a method of inserting a watermark into a text document that is robust to editing a text document or to changing a text format, that is, not easily broken by such editing or modification.

It is still another object of the present invention to provide a method for extracting a watermark even when a clustering error occurs in a text document into which a watermark is inserted.

Another object of the present invention is to provide a method of inserting and extracting a watermark that is not visually identifiable in a text document and which can prove copyright or ownership, thereby providing a variety of text documents and other networks used on the web or in electronic documents. This is to protect the copyright and ownership of the text document being shared.

1 is a diagram illustrating a conventional watermark embedding example in which a watermark is inserted into a text document by changing the space between letters.

2 is a block diagram showing the configuration of a watermark embedding apparatus according to an embodiment of the present invention.

Figure 3a is a Basic Latin table used in Unicode, Figure 3b is a Number form table used in Unicode.

4 is a diagram illustrating generation of a spread spectrum watermark.

5A is a diagram illustrating an example of a text document having a watermark inserted according to an embodiment of the present invention using ASCII code, and FIG. 5B is a text having a watermark inserted according to an embodiment of the present invention using Unicode. It is a figure which shows an example of a document.

6 is a block diagram showing the configuration of a watermark extraction apparatus according to an embodiment of the present invention.

7 is a diagram illustrating restoration of a spread spectrum watermark.

Explanation of symbols on the main parts of the drawings

100: watermark insertion device 110: target document

120: parsing unit 130: region selection unit

140: watermark information 150: error correction code

160: expansion unit 170: key insertion unit

180: random number generation unit 190: XOR conversion unit

200: shuffling portion 210: combination portion

220: Document embedded with watermark 300: watermark extractor

310: parser 320: region selector

330: reverse shuffling unit 340: random number generation unit

350: XOR conversion unit 360: contraction unit

370: watermark extraction unit 380: error correction

390: extracted watermark

In order to achieve the object of the present invention as described above, the present invention provides a method for embedding a watermark in a text document, the method comprising: generating a watermark bit string; And inserting a watermark in the text document by using the same character among characters in the text document but having different character code values.

Here, the step of inserting the watermark is set to the first code value and the second code value of the character having the same character but different code values as 0 value and 1 value of the watermark bit string, respectively, and then the text document It is preferable that the step of inserting a watermark in the text document by adjusting the code value of the character that is the same character but different in the code value.

In addition, it is preferable that the same character but different from each other in the code value is a space character of English and Korean in an ASCII code and a 2-byte complete or 2-byte combined code.

In addition, it is preferable that the same character but different in code value is an English character having the same space character or character but having different code values.

Further, it is preferable that the same character but different in code value is a character satisfying a predetermined frequency of use in the text document.

In order to achieve another object of the present invention, the present invention provides a method for embedding a watermark in a text document, the method comprising the steps of: (a) selecting a character among the characters in the text document, but different characters from the character code value; (b) generating a binarized watermark bitstream signal; (c) extending the watermark bitstream signal using a multiple of R times; (d) generating a wideband binary pseudo random number signal from a given user key; (e) generating a spread spectrum watermark bitstream signal by performing an XOR operation on the extended watermark bitstream signal and the generated wideband binary pseudorandom number; And (f) inserting a watermark into the text document using the same characters selected in the step (a) but having different character code values according to the spread spectrum type watermark bit string signal. It provides a method to include.

The method may further include (e1) shuffling the watermark bit string signal in the spread spectrum form after the step (e) but before the step (f).

In addition, the shuffling is preferably made using a random number having a predetermined seed (Seed) value.

Further, before the step (a), it is preferable to further include the step of converting the text document into a binary text document.

In addition, after the step (b) and before the step (c), it is preferable to further include (b1) adding an error correction code to the created binary watermark bit string signal.

In order to achieve another object of the present invention, the present invention provides a method for extracting a watermark from a text document having a watermark embedded by the method of embedding the watermark, (g) the watermarked text Converting the document into a binary text document; (h) finding a character string in which the code value is substituted in the binarized text document and generating a spread spectrum type watermark bit string signal from the character string in which the code value is substituted; (i) generating a wideband binary pseudo random number signal from a given user key; (j) generating an extended watermark bitstream signal by performing an XOR operation on the spread spectrum type watermark bitstream signal and the generated wideband binary pseudorandom number; And (k) contracting the extended watermark bitstream signal to restore the original watermark bitstream signal.

Here, it is preferable to further include the step (h1) performing a reverse shuffling on the generated watermark bit string signal in the form of the added peak spectrum after the step (h) and before the step (i).

In order to achieve another object of the present invention, there is provided an apparatus for embedding a watermark in a text document, comprising: means for generating a watermark bit string; And means for embedding a watermark in the text document using characters that are the same among the characters in the text document but differ in the character code values.

According to another aspect of the present invention, there is provided an apparatus for embedding a watermark in a text document, the apparatus comprising: an area selecting unit for selecting a character among the characters in the text document but having different character code values; A watermark generator for generating a binarized watermark bit string signal; An expansion unit for expanding the watermark bit string signal using a multiple of R times; A random number generator for generating a wideband binary pseudorandom signal from a predetermined user key; An XOR converter configured to generate a spread spectrum watermark bitstream signal by performing an XOR operation on the extended watermark bitstream signal and the generated wideband binary pseudorandom number; And a combination unit for inserting a watermark into the text document by using the same characters selected by the area selection unit but having different character code values according to the spread spectrum type watermark bit string signal. to provide.

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

An embodiment of the present invention can be broadly divided into a watermark embedding process of inserting a watermark into a text document and a watermark extracting process of extracting a watermark from a text document into which the watermark is inserted. First, referring to FIGS. 2 to 5. The watermark embedding process will now be described.

2 is a block diagram showing the configuration of a watermark embedding apparatus according to an embodiment of the present invention. Referring to FIG. 2, the watermark embedding apparatus 100 according to the present invention includes a parser 120, an area selector 130, watermark information 140, an error correction code 150, and an expander 160. ), Key insertion unit 170, random number generation unit 180, XOR conversion unit 190, shuffling unit 200 and combination unit 210, these components may be a hardware or software module. .

First, the target document 110 shown in FIG. 2 is a text document provided in the form of a file online, and is edited using conventional word processing software such as Notepad and WordPad, Microsoft's Word, Korean, Hunminjeongeum A document or document edited using Unicode-based word processing software.

As will be described in detail with reference to the following description, in the embodiment of the present invention, the code values are different from each other by using ASCII code and a character code value such as a two-byte completion type or a two-byte combined code or Unicode. The watermark is inserted using the same characters).

Therefore, prior to describing a series of processes performed by the parser 120 and the region selector 130 according to an embodiment of the present invention, first, the general description of the character code is the same as the character, but the code values are different from each other. A brief description will be given of the characteristic method of the present invention for inserting a watermark using different characters.

A character code is a language used by a human being that assigns a combination of bits to a specific character for communication between a computer and a human, a computer and a computer, and a computer and a peripheral device. For example, a binary code, ASCII code, 16-bit code and Unicode are well known. Characters are coded according to the language and structure of the character codes used, and these characters are prescribed by the international standard ISO / IEC2022, etc., and the structure of the character codes and the extension method of the codes are prescribed.

In Korea, two-byte completion or two-byte combinations have been used to represent Hangul. By the way, the combination type takes up less space, but there is a possibility of causing a collision with the communication control code in the communication between the interfaces, and there is a disadvantage that the shape of the font is not good. In addition, in the case of the completed type, various Hangul expression methods are difficult and occupy a lot of space.

In order to overcome the shortcomings of the combined and complete code, since 1995, Unicode has been released through the ISO10646 revision. Through this, 11172 Korean characters are supported. This Unicode has the advantages of the existing combined and complete types. The array also has a perfectly aligned code system.

Unicode has become an international standard that is used not only in non-English countries (Japan, China, Arabs, etc.) but also in English-speaking countries, and almost all languages currently used on computers are expressed using Unicode.

In the present invention, based on such an ASCII code and a two-byte completion type or two-byte combined code or Unicode, a character having the same character but different code values is selected and the watermark is inserted into the text document using the selected character. Done. Hereinafter, a description will be given in detail of selecting characters having the same characters but different code values.

First, the method based on ASCII code is demonstrated. The ASCII code was developed in 1963 for the storage and exchange of data by the American National Standards Institute (previously known as ANSI), with a total of 128 control characters, spaces, and 94 print and delete characters. This code was developed after allocating two bit combinations and became the 1968 standard.

The space of the text document has the ASCII code value of 20H. Even if this part is replaced with a space character having a different code value, the sentence remains the same. Can be inserted.

In the ASCII code, the character substituted for the watermark insertion of the present invention may use a control code in the case of a text document, and in the case of using the control code, a binary value may be assigned to each control code. In ASCII code, 32 characters from the initial 0 to 31 are not used in the actual text, so you can substitute them for the space character used in the text sentence.

For example, by replacing the ASCII code value of 20H with 00H (the null character), the user feels it the same as the space, but the code value of the character actually changes. In addition to 00H values, it is possible to assign binary codes 000B to 111B for each. In this case, in order to insert the information 01001001010011010101011B (that is, the watermark information), it is possible to insert up to 3 bits of information in one character by using the codes 02H, 02H, 02H, 04H, 06H, 05H, 02H, 03H. Do. In this method, the number of bits that can be represented by one piece of information can be determined according to the amount of replaceable character codes.

5A is a diagram showing an example of a text document in which a watermark is inserted according to an embodiment of the present invention using an ASCII code. Referring to FIG. 5A, the space character before the watermark is inserted is replaced with the control character after the watermark is inserted. However, although the watermark is inserted according to a predetermined rule in which the code value of the space character is different and this code value is different, the text document does not visually make a difference at all. Such an effect cannot be achieved in the conventional open space method or the method of inserting a watermark by changing the font. As described above, the method based on the ASCII code of the present invention can be equally applied to text documents written in Korean as well as English characters.

The method of using the English and Korean space characters in the ASCII code as described above is equally applicable to a two-byte completion type or a two-byte combined code.

Next, a method based on Unicode or extended ASCII code will be described. In Unicode or extended ASCII code schemes, the same character may have two code values. If the bit value of the information to be inserted by selecting such a character is 0, the code value of the originally used area is used as it is. If the bit value of the information to be inserted is 1, the code value of another area is used. will be. In this case, the replaced one code may be a method of expressing only one bit, and thus, a watermark is inserted according to information inserted as 0 or 1.

Figure 3a is a Basic Latin table used in Unicode, Figure 3b is a Number form table used in Unicode. Referring to FIG. 3A, since the C0 Controls and Basic Latin table is configured in the same form as the ASCII code, the watermark can be inserted by replacing the space character in the ASCII code as described above.

In addition, by using the characters used in the number forms table of FIG. In the tables of FIGS. 3A and 3B, the watermark information can be inserted without any change in the content of the text when the 14 characters having the same but different code values are used.

5B is a diagram illustrating an example of a text document having a watermark inserted according to an embodiment of the present invention using Unicode. 5B shows the result of replacing all the strings of the basic Latin code used as the alphabet in the text characters using Unicode with the strings of the number forms code. Referring to FIG. 5B, for example, characters such as 'M', 'c', and 'v' have been replaced, but this is negligible to the naked eye. As shown in FIG. 5B, the string to be replaced can be not only for all the target strings but also for a part of the string, or for the use of a space character and a combination of the space character and the target string.

Referring back to FIG. 2, the content of the text provided from the target document 110 is converted into a binary text form through the parser 120 to record the number and position of the characters.

Then, the area selection unit 130 selects a predetermined character among the characters checked by the parser 120 having a high frequency of use and being replaced with a character having the same character but having a different code value. Substitutable characters herein refer to characters that are the same as described in the ASCII code or Unicode described above, for example, but whose code values are different from each other. The selected target character may be two or more characters depending on one or the characteristics of the target document.

In addition, the selected target character needs to select a character with a high frequency of use, that is, a predetermined frequency or more, since it is necessary to secure a space that can be inserted in order to insert a certain amount of watermark. The larger the insertion space is, the more watermark information can be inserted, which is closely related to the robustness of the watermark or the amount of information.

Next, a process of generating a watermark including information such as copyright, ownership, and user to be inserted into a text document will be described.

The watermark information 140 is user information inputted by using an input device such as a keyboard or preset user information, or information which can prove copyright or ownership. The watermark information is converted into a binary bit string.

4 is a diagram illustrating generation of a spread spectrum watermark, and FIG. 4A corresponds to watermark information converted into the form of the binarized bit string.

Next, an error correction code 150 is added to the watermark information converted into the form of a binarized bit string. For example, a parity bit or a CRC may be used as the error correction code. The error correction code used herein serves to correct an error that may occur in the above process through an error code.

In the expansion unit 160, the number of sampling is increased by using a multiple of R times the binarized bit string information A, which is represented as shown in FIG. Since the converted bit string contains the information of the user to be inserted into a single unit, the number of samplings is expanded to R by increasing the number of samplings, thereby preventing the user information from being damaged by editing and extracting the watermark more easily. .

Here, R is determined by the frequency of the characters selected by the area selector 130. That is, if the user information is composed of N bit strings, and the frequency in the text document of the selected character is M, then an R value having a range satisfying the expression M > N * R is selected.

On the other hand, a method of expanding the binarized bit stream by R times is, for example, a method of repeatedly R times the entire binarized bit stream. That is, if the two-evolved bit string is '011101', when R is 3, the bit string whose sampling number is increased by R times becomes '000111111111000111'.

The key inserting unit 170 is a device for generating a key for generating a random number, so that the user can arbitrarily adjust user-defined information or a seed value, thereby improving security. The random number generator 180 receives random key information from the key inserter 170 and generates random numbers. The random number generation method used here may be a transformation method, a rejection method, a Monte Carlo method, or any similar method. As described above, the binary pseudorandom number generated by the specific key is shown in Fig. 4C.

Next, the XOR conversion unit 190 performs an XOR (exclusive OR) operation on the extended watermark bit string (B) from the expansion unit 160 and the wideband binary pseudorandom number (C) from the random number generation unit 180. Generate a bit string watermark (shown in FIG. 4D) in the form of spread spectrum.

The bit stream type watermark generated in the form of spread spectrum can restore the original information even if a part of the watermark is damaged or manipulated, and the watermarks of the different watermarks according to the specific key used to generate the binary pseudorandom number can be recovered. It can be inserted to maintain security and use the assigned key for each individual to insert a watermark.

Next, the watermark shuffling process performed by the shuffling unit 200 will be described.

Since the watermark bit string generated by the XOR converter 190 is generated by an oversampling scheme, repetition of the same bit string appears, and if more than R transformations occur, information in a predetermined region is damaged and thus the watermark information is changed. Restoration can be difficult. That is, in order to prevent watermark extraction from being impossible due to the damage of watermark information of a certain region due to a burst error, the repeated watermark bit strings are mixed and rearranged so that even if a cluster error occurs, It is possible to extract the watermark later by preventing the mark bit string from being intensively damaged.

As a method of mixing the repeated bit strings in the shuffling unit 200, both a random order sequence and a random sequence sequence may be used. Since a watermark bit string in the form of a shuffled spread spectrum is generated, original information can be recovered even if a portion of the watermark is damaged or manipulated, and different watermarks are generated according to a specific key used to generate a binary pseudorandom number. Can be inserted to maintain security, and the key assigned to each individual to insert a watermark can be used. Such a method is called a direct sequence spread spectrum method, and a method of randomly changing a carrier frequency is called a frequency hopping spread spectrum method.

In addition, as one of the spread spectrum methods, a random number generated by a random number generator is inserted as a watermark into an original document, and the correlation is obtained at the time of detection. When the correlation value exceeds a predetermined reference value, a watermark exists. If not, a method of determining that there is no watermark is possible.

Since the seed value in the method using the random number sequence also exists as a key of the watermark, a person who does not know the seed value may have additional security and stability because it is not easy to insert or extract the watermark.

Referring to FIG. 4 again, the above-described process will be described. A portion in which a signal composed of N bits is elongated in FIG. When the binary number is '0', the watermark bit string to be inserted may be represented as '111111100000000'.

Multiply R by '1 * R 1 * R 1 * R 1 * R 1 * R 1 * R 1 * R 0 * R 0 * R 0 * R 0 * R 0 * R 0 * R 0 * R 0 * R When expanded in the form of ', it becomes an oversampled watermark bit string as shown in FIG.

As a result of the XOR operation of the extended watermark bit string (FIG. 4B) with the wideband binary pseudorandom number (C) in FIG. 4, the water having a spread spectrum form as shown in FIG. The mark bit string is generated.

Since the watermark bit string signal in the spread spectrum form is a repeating form of a constant signal string, the shuffling unit 200 performs a mixing process using a random number sequence in order to encrypt it.

The generated watermark bit string in the form of spread spectrum is inserted in the combining unit 210 at a predetermined position of the target character selected by the region selecting unit 130. That is, the syntax analyzing unit 120 and the region selecting unit. The watermark is inserted by adjusting the code value of the target character having the same character but different code values and having a predetermined frequency determined in step 130 according to the watermark bit string value of the spread spectrum form.

That is, for example, in the case of Unicode, if the bit value of the watermark to be inserted is '0', the target character has the code value used originally, and if the bit value of the watermark to be inserted is '0', Have the character have a replacement code value.

In addition, the combination unit 210 performs a function of converting a document in the form of a binary text having a watermark inserted therein into a text form in a form usable by the user again. Is generated.

As described above, the document 220 in which the watermark is inserted is as shown in FIG. 5A or 5B, for example, and the character string thus displayed cannot be detected by a general text viewer program or a word processing program. The inserted watermark information is not affected even if the font, size, sentence structure, etc. of the character string are converted during editing. In addition, even when a part of the text is edited, the insertion information of the watermark is not affected because the entire insertion information of the watermark is not affected.

Next, a process of extracting a watermark from the document 220 into which the watermark is inserted will be described with reference to FIGS. 6 and 7.

6 is a block diagram showing the configuration of a watermark extracting apparatus according to an embodiment of the present invention, and FIG. 7 is a diagram showing restoration of a spread spectrum watermark.

Referring to FIG. 6, the watermark extraction apparatus 300 of the present invention includes a syntax analyzer 310, an area selector 320, an inverse shuffling unit 330, a random number generator 340, and an XOR converter ( 350, a constrictor 360, and a watermark extractor 370, these components may be hardware or software modules.

In the parsing unit 310, the watermark-embedded document 220 is converted into a binary string, and the area selection unit 320 determines whether the frequency of use and the code value are replaced with respect to the target character into which the watermark is inserted. Check whether or not. If there are no characters whose code values are substituted for the target character, the result is that the watermark information is not inserted in the text document.

In the reverse shuffling unit 330, a reverse process of the mixing process performed in the shuffling unit 200 of the watermark inserting apparatus 100 is performed. That is, if shuffled by random sequence method, the random number sequence is regenerated by key value to restore the shuffled spread spectrum type watermark bit string in the original order, and if shuffled according to the predetermined order, according to the original order. Restore The watermark bit string in the form of spread spectrum reversely shuffled by the inverse shuffling unit 330 is shown in FIG. 7A.

In the random number generator 340, a wideband binary pseudorandom number is generated according to a given key value and this bit string is shown as shown in FIG. The wideband binary pseudorandom number generated by the random number generator 340 is a bit string signal such as the wideband binary pseudorandom number generated by the random number generator 180 of FIG. 2.

In the XOR conversion section 350, the spread spectrum type watermark bit string (A in FIG. 7) output from the inverse shuffling section 330 and the wideband binary pseudo-random number output from the random number generator 340 (B in FIG. 7). ) Is XORed, and the resulting extended watermark bitstream signal is shown as C in FIG.

Next, the constrictor 360 reduces the expanded watermark bitstream signal (C of FIG. 7), and restores the original watermark bitstream signal (D of FIG. 7) by the watermark extractor 370. do.

Next, the reconstructed watermark bit string signal is obtained by the watermark 390 extracted through the error correction process 380, thereby obtaining copyright or ownership information of the text document inserted by the watermark embedding apparatus 100. It can be seen and can determine whether the text document has been forged or tampered with.

Although the invention has been particularly shown and described with reference to the above embodiments, it has been used for the purpose of illustration and those of ordinary skill in the art to which the invention pertains have the spirit and scope of the invention as defined in the appended claims. Various modifications can be made without departing.

As described above, according to the present invention, there is an effect that a watermark having copyright or ownership information can be inserted without making a visually identifiable change to a sentence or word in a text document.

In addition, according to the present invention, it is effective to insert a watermark into a text document, which is robust to editing the text document or changing the text format, that is, not easily broken by such editing or changing.

In addition, according to the present invention, even if a clustering error occurs in the text document into which the watermark is inserted, the watermark can be extracted.

In addition, the present invention provides a method of inserting and extracting a watermark that cannot be visually identified in a text document and which can prove copyright or ownership, thereby sharing on a variety of text documents and other networks used on the web or in electronic documents. The effect is to protect the copyright and ownership of the text document being created.

Claims (16)

In the method of embedding a watermark in a text document, Generating a watermark bit stream; And And embedding a watermark in the text document by using the same characters among the characters in the text document but having different character code values. The method of claim 1, Inserting the watermark The first code value and the second code value of the character having the same character but different code values are set to 0 and 1 value of the watermark bit string, respectively, and then the character having the same character but different code values in the text document And embedding a watermark in the text document by adjusting a code value of the text document. The method according to claim 1 or 2, The same character but different in code value is a space character of English and Korean in ASCII code and 2-byte complete or 2-byte combined code. The method according to claim 1 or 2, The character of the same character but different code values is a space character of the English and Korean characters in Unicode or the same character, but characterized in that the code values are different from each other. The method according to claim 1 or 2, And said characters having the same characters but different code values are characters satisfying a predetermined frequency of use in said text document. In the method of embedding a watermark in a text document, (a) selecting a character among the characters in the text document but having different character code values; (b) generating a binarized watermark bitstream signal; (c) extending the watermark bitstream signal using a multiple of R times; (d) generating a wideband binary pseudo random number signal from a given user key; (e) generating a spread spectrum watermark bitstream signal by performing an XOR operation on the extended watermark bitstream signal and the generated wideband binary pseudorandom number; And (f) embedding a watermark in the text document using the same characters selected in step (a) but having different character code values according to the watermark bit string signal in the spread spectrum form Characterized in that. The method of claim 6, Step (f) The first code value and the second code value of the character having the same character but different code values are set to 0 and 1 value of the watermark bit string, respectively, and then the character having the same character but different code values in the text document And embedding a watermark in the text document by adjusting a code value of the text document. The method according to claim 6 or 7, After step (e) and before step (f), (e1) shuffling the watermark bitstream signal in the form of a spread spectrum. The method of claim 8, The shuffling is performed using a random number having a predetermined seed value. The method of claim 6, Before step (a) above, Converting the text document into a binary text document. The method of claim 6, After step (b) and before step (c), (b1) adding an error correction code to the created binary watermark bit string signal. A method of extracting a watermark from a text document having a watermark embedded therein by the method of claim 6, (g) converting the text document containing the watermark into a binary text document; (h) finding a character string in which the code value is substituted in the binarized text document and generating a spread spectrum type watermark bit string signal from the character string in which the code value is substituted; (i) generating a wideband binary pseudo random number signal from a given user key; (j) generating an extended watermark bitstream signal by performing an XOR operation on the spread spectrum type watermark bitstream signal and the generated wideband binary pseudorandom number; And (k) contracting the extended watermark bitstream signal and restoring it to the original watermark bitstream signal. The method of claim 12, After step (h) and before step (i), (h1) further comprising performing a reverse shuffling on the generated spread spectrum type watermark bit string signal. An apparatus for embedding a watermark in a text document, Means for generating a watermark bit stream; And And means for embedding a watermark in the text document using characters that are the same among the characters in the text document but differ in character code values. An apparatus for embedding a watermark in a text document, An area selection unit for selecting characters from among the characters in the text document, but characters having different character code values; A watermark generator for generating a binarized watermark bit string signal; An expansion unit for expanding the watermark bit string signal using a multiple of R times; A random number generator for generating a wideband binary pseudorandom signal from a predetermined user key; An XOR converter configured to generate a spread spectrum watermark bitstream signal by performing an XOR operation on the extended watermark bitstream signal and the generated wideband binary pseudorandom number; And And a combination unit for inserting a watermark into the text document by using the same characters selected by the region selection unit but having different character code values according to the spread spectrum type watermark bit string signal. Device. The method according to claim 14 or 15, The embedding of the watermark is the same character in the text document after setting the first code value and the second code value of the character having the same character but different code values to 0 and 1 value of the watermark bit string, respectively. And adjusting the code values of the characters having different code values.