EA044732B1 - METHOD AND SYSTEM FOR PROTECTING INFORMATION FROM LEAKAGE WHEN PRINTING DOCUMENTS USING THE IMPLEMENTATION OF DIGITAL MARKS - Google Patents

Description

Field of technology

The declared solution relates to the field of information security, in particular to solutions for preventing information leakage when printing documents.

State of the art

Leak prevention technologies (English: Data Leak Prevention, DLP) are technologies for preventing leaks of confidential information from an information system to the outside, as well as technical devices (software or firmware) for such leak prevention.

From patent application US 20080091954 Al (Morris et al., 04/17/2008) a solution is known for checking the integrity of data presented on printed documents. The solution is based on the use of a unique identifier, which is used to analyze the contents of the document. Each segment of a document is assigned a digit or group of digits, and each page or segment of a document can be assigned a single digit in the common identifier. The collection of digits associated with the document is combined into an authentication string. When a request for subsequent document processing is received, authentication and document integrity are verified by reading the submitted document to obtain an authentication string, and then comparing the new string with the previously stored string. Once successfully matched, the document is considered valid, authenticated, and unaltered.

The disadvantage of this solution is the impossibility of using it to prevent leaks in order to identify the employee who committed the leak when printing documents. Also, another disadvantage is the insufficient effectiveness of document protection, which is due to the use of a code to compare the authenticity of a document, which only allows one to establish the fact of the immutability and authenticity of the document, but does not prevent information leakage.

The essence of the invention

The claimed invention is aimed at solving a technical problem, which is to create an effective means for protecting digital information from leakage during printing.

The technical result consists in increasing the efficiency of data protection from leakage by introducing digital tags into the document that encode a unique user identifier for subsequent identification when analyzing printed documents.

The claimed result is achieved through a method of encoding information to protect against leaks when printing documents, performed using the processor of a computer device, and the method contains the stages of:

receiving on the user's computer device information about printing at least one digital document containing at least text, wherein the computer device is associated with a unique identifier (UID) of the user;

Before the digital document is sent for printing, it is processed, during which the letters contained in the digital document are recognized;

encoding the user's UID into a set of digital marks that are located on the outlines of the letters and/or near the outlines of the letters of the digital document;

transmitting a digital document for printing with an encoded user UID.

In one of the particular examples of implementation of the method, digital document recognition is performed using optical character recognition (OCR). In another particular example of the implementation of the method, all characters on each page of a digital document are recognized.

In another particular example of implementing the method, each user UID character is encoded into binary code.

In another particular example of implementing the method, the area of placement of digital marks is determined based on the bit of the binary code.

The claimed technical result is also achieved by implementing a method for protecting information from leaks on printed documents, performed using a processor of a computer device, wherein the method contains the steps of:

obtaining at least a portion of the image of the printed document with the user UID encoded in the above manner;

perform recognition of the resulting image;

identify letters containing digital marks in their vicinity;

performing determination and extraction of the encoded UID.

In one of the particular examples of the method, digital document recognition is performed using OCR.

The claimed solution is also implemented using corresponding systems containing a processor and memory that store machine-readable instructions for implementing each of the above methods.

Brief description of the figures

Fig. 1 illustrates a flowchart of a digital mark encoding method.

Fig. 2A - 2B illustrate examples of the placement of digital marks in a digital document.

- 1 044732

Fig. 3 illustrates a block diagram of digital mark decoding.

Fig. 4 illustrates a diagram of the hour of disclosure of UID positions.

Fig. 5 illustrates a general view of a computing device.

Carrying out the invention

In fig. 1 presents a method (100) for protecting information in digital documents from leakage by encoding the user UID in the form of digital marks into the document. At the first stage (101), information about printing of the digital document is obtained. The method (100) is carried out on a computer device of a user, for example, an employee, and a user UID is associated with the device, allowing him to be identified. The execution of step (101) is animated by software logic executed by a computer device and can be implemented, for example, in the form of a software agent or module that provides signals from the processor indicating that a digital document is being sent for printing. A digital document is typically a file and can contain text, graphics, or a combination of both.

After receiving a command on the device to intercept and analyze the document before sending it to the printer, at step (102) recognition of the mentioned digital document is performed. Document processing is done using OCR technology to ensure recognition of letters and symbols in a digital document.

After the digital document recognition step, the UID encoding process is carried out at step (103). The UID is, for example, a numeric personnel number of an employee - a digital TAB code, consisting, for example, of 8 digits. This code can be represented as an array of digits TAB _v = {η _1Λ η ₂ , -n _t ], TAB ₈ £ [0 ...9], t = 8.

A schematic view of the code is presented in table. 1.

Table 1 Schematic representation of the personnel number

tlv ₈ Number ^P 1 p ₄ ^p 5 ^p b p ₇ ^By Position 1 2 4 5 6 7 8

Each element of the personnel number is a number from 0 to 9, respectively, each element of the personnel number can be displayed in binary form with a dimension of 4 bits, i.e. it will represent a binary number from 1 to 1100, which is a homomorphism with a shift, presented in Table 2.

table 2

Scheme of homomorphism of personnel number from decimal to binary number system

tav£ ^es TAB^ ⁱⁿ 0 0001 1 0010 2 UN 3 0100 4 0101 5 it 6 0111 7 1000 8 1001 9 1010

The display of 0 in 0001 is necessary in order to record the presence of 0 in the personnel number. To encode a personnel number element in binary code TAB ₈ ^IN = [b _lf b ₂ ,b ₃ , ..., b),1 = 8, 4 digits are required ~ i ^c i' ^c 2 > ^c z > ^c 41, example which are presented in table. 3.

Table 3

Schematic division of a binary number into digits

Thus, it is possible to encode any number into a letter using binary encoding.

An example of such a division for subsequent encoding is presented in Fig. 2A - fig. 2B. Each recognized letter (20) is divided into 4 quarters in a clockwise plane, starting from the lower left corner.

If there is a 1 in the 1st digit of the binary representation of the personnel number digit, _{the 1st} mark is placed in the 1st quarter. Similar operations are carried out with all bits of the binary representation of a digit.

- 2 044732

The method of applying a mark in the space near the letter is that, as shown in Fig. 2B2V a digital mark is applied in the form of a line (21) on the surface of the letter or a point (22) in the vicinity of the letter in a given quarter. An example of encoding marks into letters is presented in table. 4.

Table 4

Position coding scheme

The above table 4 means that each number position in the personnel number can be encoded into any of the 4 letters. The selection of letters for marking is carried out page by page. Let document D contain 1 pages, then document D is an array of pages, “ZhtRgAz - Ρι}> I ^e

On each page Р^ * ^G Ж the text is read character-by-character and written into an array of characters = {^,52,53,...5, }, , ^ρ ι where l _pi is the number of characters on page pi, from which Russian letters are identified ^S Pi ^:

Wrus, kvy ¹ 'Pi ^{e S} PiNext, 8 arrays P° ^s i>P° ^s 2 -Pos ₈ are created, each of which corresponds to each position of the personnel number. Each Pos array is filled with those characters from Wruspi that correspond to the position from Table 4. For example, Pos ₁ is filled with all characters from Wruspi that have the values {a, z, p, h}, regardless of case.

The arrays P° ^s i>P° ^s 2 -Pos ₈ are mixed, for example, by Knuth shuffle.

Let Iposp ¹ pos ₂ > lpos ₃ - hos ₈ - the dimensions of the resulting arrays, P - the percentage of characters for the implementation of the label ^{p e} [°' ³ -°' ⁷ ]' then each array from P° ^S ^P° ^S 2 -P° ^s 8 is cut from end to size

Σ^ογ ' Λ Pos ₁ ,Pos ₂ ...Pos _B

Pos ^p ,Pos ₂ ...Posq, π PnF Reach? ι

The resulting arrays ^Γυ ^ΐ' ^ΓυΛ 2 -™s are used to apply digital marks in the manner described above. Digital tagging is done by cutting out letters using OCR, adding the tagging to pixel coordinates, and adding the digitally tagged letters back into the document to be printed. After embedding all the labels (21, 22) on the desired page pi, the same is done for the next page p _i+1 and so on until the end of the document p _b

In table Figure 5 shows an example of coding labels for user UID - 00013400.

Table 5

An example of encoding digital marks in the neighborhood of letters

- 3 044732

After digital labels encoding the UID are added to the document sent for printing, at step (104) it is sent for printing. The printed document will contain an encoded UID that is indistinguishable to the human eye. The size of digital marks can be chosen arbitrarily (for example, marks with a radius of 1 - 2 pixels).

In fig. 3 shows the sequence of steps performed when performing the method (300) for recognizing UID on printed documents. At step (301), the computing device used to determine the UID in the printed document obtains an image of such document. The image may contain all or part of the text, with an encoded UID, obtained, for example, by photographing with an external device (smartphone, camera, etc.) or by scanning a printed document using OCR.

Next, at step (302), also using OCR technology, letters in the document are recognized, and if there are several pages in the document, then each page of the document is recognized. At step (303), digital marks in the vicinity of the recognized letters are read. An example of analyzing digital tags can be carried out according to the example given in table. 5, which can be used as a table to match labels to the corresponding digit of the user UID. After this, the UID is decoded at step (304) and the personnel number of the employee and the corresponding user from whose computer device the document was printed is determined.

Mathematical justification of the method.

To do this, we will make sure that the frequencies of opening positions ^_ { ^η ι> ^η 2, - ⁿ t}> ^m - g are uniformly distributed for all m, which allows us to show the probability of extracting a personnel number (UID) from the text of the page.

For mathematical justification, a study was carried out on the frequency of occurrence of letters in text with different contents; for example, let’s consider such a distribution typical for literary works. List of literary works participating in the experiment: The Silmarillion. J.R.R. Tolkien, Twenty Thousand Leagues Under the Sea. Jules G. Verne, Twenty Years Later. Alexandre Dumas, Three Musketeers. Alexandre Dumas, Gone with the Wind. Margaret Mitchell, Ivanhoe. Walter Scott, Hero of Our Time. N.V. Gogol, War and Peace. L.N. Tolstoy, Inhabited Island. Boris and Arkady Strugatsky, Crime and Punishment. F.M. Dostoevsky, The Living and the Dead. K.M. Simonov, total 8,366,594 characters, 3919 pages. Mathematical linguistics has shown the following probabilities for the frequency of occurrence of letters of the Russian alphabet in texts (Table 6).

Table 6

Table of frequency of occurrence of letters of the Russian alphabet in fiction

Letter Meeting frequency, % Letter Meeting frequency, % A 8.31 R 4.32 b 1.65 With 5.24 V 4.59 T 6.06 G 1.72 U 2.95 d 3.06 F 0.13 e 8.42 X 0.84 e 0.02 h 1.56 and 1.01 ts 0.44 3 1.71 w 0.97 And 6.84 sch 0.32 th 1.11 ъ 0.03 To 3.42 s 1.81 l 4.99 b 1.93 m 3.16 uh 0.27 n 6.46 Yu 0.57 O 11.42 I 1.95

n 2.71

- 4 044732

To obtain the value of the position opening frequency ^_ { ^η ι> ^η 2, - ⁿ t} the following actions are performed. From the table 4 and 5, the letters into which the digits are encoded are known. To obtain the frequency of bit opening for the algorithm for applying a mark in the space near a letter, the frequencies of the letters into which the marks are encoded are added up, because The position is opened when a mark is detected in at least one of them. As a result of the above actions, a table is obtained. 7.

Table 7

Table of frequency of disclosure of positions of personnel number i Frequency of letters Frequency of disclosure of digits

A 8.31 3 1.71 And 2.71 h 0.44 13.17 b 1.65 And 6.84 R 4.32 w 0.97 13.78 V 4.59 th 1D1 With 5.24 SCH 0.32 11.26 G 1.72 To 3.42 T 6.06 ъ 0.03 11.23 d 3.06 l 4.99 at 2.95 s 1.81 12.81 e 8.42 m 3.16 f 0.13 uh 0.27 11.84 e 0.02 And 6.46 X 0.84 b 1.93 I 1.95 11.57 and 1.01 ABOUT 11.42 C 1.56 Yu 0.57 14.56

Based on table 7 the diagram shown in FIG. 4. The diagram shows that the frequency of opening of all positions is distributed relatively evenly.

Let's calculate the number of each letter of the Russian alphabet in the experimental sample:

Table 8

Letter-positional quantitative characteristics of the experimental sample.

Ir i Number of SYMBOLS in Sim v/p Number of OB symbols Sim v/p Number of OB symbols Sim v/p Number of OB symbols Sim v/p Number of OB symbols Sim in/page 1 A 689 971 176 3 142 242 36 P 213 560 54 h 121417 31 2 b 140 050 36 AND 572 432 146 R 372 611 95 w 74,968 19 3 V 372 447 95 th 92,969 24 With 448 533 114 SCH 26,501 7 4 G 152,827 39 To 283 925 72 T 516 921 132 ъ 2,588 1 5 D 255 254 65 l 420 003 107 U 234 845 60 s 162,890 42 6 e 709 671 181 m 257 188 66 F 12,025 3 b 165 VP 42 7 e 5,953 2 And 536 626 137 X 75 243 19 uh 24,582 6 I 171 316 44 8 and 88 798 23 ABOUT 940 740 240 C 31 199 8 Yu 51 195 13

For the method of placing a dot in the space near the letter, the following assumption is made: the percentage P of characters for the implementation of the label P = 0.3; when transmitted through messengers, a certain percentage M = 0.7 of labels is lost. Based on the above, you can calculate the probability of text recognition if the following is available for decryption:

whole page;

1/2 page;

_1/4 pages.

- 5 044732

Table 9

Explanations and probabilities of recognizing text encoded by placing a dot in the space near the letter

n _if ί 30% of the total number is selected va in rank transmission via RU messenger (70% of tags are lost) whole page 1\2 1\4 letters current position on the page POSITION recognized or not letters current position on the page POSITION recognized or not position was recognized or not POSITION recognized or not 1 89 27 27 1 13.4 1 6.7 1 2 89 27 27 1 13.3 1 6.7 1 3 72 22 22 1 10.8 1 5.4 1 4 73 22 22 1 11.0 1 5.5 1 5 82 25 25 1 12.3 1 6.2 1 6 88 26 26 1 13.1 1 6.6 1 7 49 15 15 1 7.4 1 3.7 1 8 85 26 26 1 12.8 1 6.4 1 100 100 100

Example of Experimental Application.

During testing, about 500 pages of different content were printed and analyzed: text, sparse text, text with tables, text with graphs, text with formulas;

with different types of fonts: Arial, Calibri, Times New Roman;

different text formats: regular, italic, bold, underlined;

different sizes: 12px, 14px;

different line spacing: 0.5, 1.15, 1.5;

different character spacing: regular, sparse, compacted; In each case, the possibility of extracting the label from:

direct printouts;

from a photograph of a printout;

a photo printout sent via messenger.

Printing was carried out on a Lexmark MX71 lde office black-and-white laser printer on Snegurochka office paper with whiteness CIE 146 according to ISO 11475. Photographing was done on a Samsung A51 phone in office lighting, the paper lay horizontally on the table, photographing was random at different, slight angles, about 2- 4% in 3 dimensions. When transferring photos, the Telegram messenger was used with image compression when sending.

During the experiment, parameters were selected, such as the size of the marks, their optimal locations and methods of application. The results of the last phase of the experiment are shown in table. 10.

Table 10

Experiment result

TEXT POSITION 1 2 3 4 5 6 7 8 100% detected letter A AND WITH T D e n ABOUT SPARSE TEXT POSITION 1 2 3 4 5 6 7 8 100% detected letter A AND With T L e n ABOUT TABLE POSITION 1 2 3 4 5 6 7 8 100%

detected letter P And With T D m n and SCHEDULE POSITION 1 2 3 4 5 6 7 8 detected letter A AND V T L AND FORMULA POSITION 1 2 3 4 5 6 7 8 detected letter h AND TO f n

The table described above shows good results from the analysis of photographs of printouts sent via messenger on an office black and white printer. As a result of the experiment, optimal parameters were selected for introducing marks, which, on the one hand, would be noticeable on printouts as printer defects, on the other hand, could be easily extracted from photographs sent via instant messengers.

-

Claims (6)

In fig. 5 is an overview of a computing device (500) suitable for performing the above methods. The device (500) may be, for example, a computer, a server, or other type of suitable computing device. In the general case, a computing device (500) contains one or more processors (501), memory devices such as RAM (502) and ROM (503), input/output interfaces (504), and input/output devices ( 505), and a network communication device (506). The processor (501) (or multiple processors, multi-core processor) may be selected from a variety of devices commonly used today, such as Intel™, AMD™, Apple™, Samsung Exynos™, MediaTEK™, Qualcomm Snapdragon™, etc. . A graphics processor, for example, Nvidia, AMD, Graphcore, etc., can also be used as a processor (501). RAM (502) is a random access memory and is designed to store machine-readable instructions executed by the processor (501) to perform the necessary operations for logical data processing. RAM (502) typically contains executable operating system instructions and related software components (applications, program modules, etc.). ROM (503) is one or more permanent storage devices, such as a hard disk drive (HDD), a solid state drive (SSD), flash memory (EEPROM, NAND, etc.), optical storage media (CD-R) /RW, DVD-R/RW, BlueRay Disc, MD), etc. To organize the operation of device components (500) and organize the operation of external connected devices, various types of I/O interfaces (504) are used. The choice of appropriate interfaces depends on the specific design of the computing device, which can be, but is not limited to: PCI, AGP, PS/2, IrDa, FireWire, LPT, COM, SATA, IDE, Lightning, USB (2.0, 3.0, 3.1, micro, mini, type C), TRS/Audio jack (2.5, 3.5, 6.35), HDMI, DVI, VGA, Display Port, RJ45, RS232, etc. To ensure user interaction with the computing device (500), various means (505) of I/O information are used, for example, a keyboard, a display (monitor), a touch display, a touch pad, a joystick, a mouse, a light pen, a stylus, a touchpad, trackball, speakers, microphone, augmented reality tools, optical sensors, tablet, light indicators, projector, camera, biometric identification tools (retina scanner, fingerprint scanner, voice recognition module), etc. The networking facility (506) allows the device (500) to communicate via an internal or external computer network, such as an Intranet, Internet, LAN, etc. One or more means (506) may be used, but not limited to: Ethernet card, GSM modem, GPRS modem, LTE modem, 5G modem, satellite communication module, NFC module, Bluetooth and/or BLE module, Wi-Fi module and etc. Additionally, satellite navigation tools can also be used as part of the device (500), for example, GPS, GLONASS, BeiDou, Galileo. The submitted application materials disclose preferred examples of implementation of a technical solution and should not be interpreted as limiting other, particular examples of its implementation that do not go beyond the scope of the requested legal protection, which are obvious to specialists in the relevant field of technology. CLAIM 1. A method for encoding information to protect against leaks when printing documents, performed using a processor of a computer device, wherein the method contains the steps of: receiving on the user's computer device information about printing at least one digital document containing at least text, wherein the computer device is associated with a unique identifier (UID) of the user; Before the digital document is sent for printing, it is processed, during which the letters contained in the digital document are recognized; encoding the user's UID into a set of digital marks that are located on the outlines of the letters and/or near the outlines of the letters of the digital document; transmitting a digital document for printing with an encoded user UID. 2. The method according to claim 1, characterized in that recognition of the digital document is performed using optical character recognition (OCR). 3. The method according to claim 2, characterized in that all characters on each page of the digital document are recognized. 4. The method according to claim 1, characterized in that each user UID character is encoded into binary code. 5. The method according to claim 4, characterized in that based on the bit of the binary code, the area where the digital marks are placed is determined. 6. A method for protecting information from leaks on printed documents, performed using a processor of a computer device, wherein the method contains the steps of: -