[go: up one dir, main page]

WO2023023502A1 - Procédé de marquage d'éléments visuels d'informations en vue d'une identification ou d'une authentification ultérieure - Google Patents

Procédé de marquage d'éléments visuels d'informations en vue d'une identification ou d'une authentification ultérieure Download PDF

Info

Publication number
WO2023023502A1
WO2023023502A1 PCT/US2022/075001 US2022075001W WO2023023502A1 WO 2023023502 A1 WO2023023502 A1 WO 2023023502A1 US 2022075001 W US2022075001 W US 2022075001W WO 2023023502 A1 WO2023023502 A1 WO 2023023502A1
Authority
WO
WIPO (PCT)
Prior art keywords
digital image
image
digital
comparison
file
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2022/075001
Other languages
English (en)
Inventor
Sergey VOYNOV
Alexaander KORZNIKOV
Sergei TIMOSHENKO
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
G 71 Inc
Original Assignee
G 71 Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US17/403,127 external-priority patent/US11606475B2/en
Application filed by G 71 Inc filed Critical G 71 Inc
Publication of WO2023023502A1 publication Critical patent/WO2023023502A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
    • G06V10/225—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition based on a marking or identifier characterising the area
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/95—Pattern authentication; Markers therefor; Forgery detection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition

Definitions

  • the invention is related to the field of information visuals (visual images) marking for their subsequent identification or authentication (establishing the fact of authenticity) that can be used for images protection from unauthorized disclosure (identification) or for prevention of fraud/establishing the fact of authenticity (authentication).
  • the proposed method is devoid of this disadvantage and can be applied to any visualized information (on the screen or paper), while retaining all the advantages.
  • the proposed method does not rely on words in a document, but operates with parts of an image combined into groups.
  • Visualized information can be understood as a regular document, as well as a diagram or a drawing, a table or presentation, a web interface or a native GUI, etc.
  • the given invention helps to provide image protection from unauthorized spread, identification of distribution channels (persons involved in distribution), as well as can be used to confirm the authenticity of the image.
  • the image in the framework of this invention means any graphic image (digital on the screen or displayed on a printing device), but the main use scenario of the invention involves marking images of documents (pages), diagrams, drawings, presentations, tables, but the invention is not limited to any particular type of images. Using the method of marking with the above types of images allows to achieve such an advantage as the indistinguishability (invisibility) of marking "by eye” without the use of special technical means.
  • One exemplary embodiment of the inventive method for authenticating digital information includes obtaining, in digital form, information for authentication; preparing the information for processing, such preparation comprising converting the information into a digital image; identifying segments of content in the digital image; grouping the segments of content into one or more segment groups; generating a marking sequence comprising shifting at least one of the one or more segment groups in one or more directions; and applying the marking sequence to the digital image, creating a unique marked copy of the digital image.
  • the marking sequence further includes altering a spacing between two or more of the segments in at least one of the one or more segment groups.
  • the information for authentication may be obtained via electronic communication from a user. Additional data relating to the user and the user’s submission may also be obtained along with the information for authentication. For example, the additional data relating to the user and the user’s submission may include identifying information about the user, a date and time of the user’s submission, and the user’s method of submission. At least one of the marking sequence, the unique marked copy, the digital image, the information for authentication, and the additional data relating to the user and the user’s submission may be stored in digital form in a database.
  • the method also includes obtaining a digital image for comparison; employing an authentication process to authenticate the digital image for comparison by comparing it to the unique marked copy; and generating an authenticity conclusion report identifying the digital image for comparison as authentic if the authentication process determines that it matches the unique marked copy and identifying it as inauthentic if it does not.
  • the authentication process may be performed by digitally overlaying the digital image for comparison with the unique marked copy, and the digital image for comparison may be obtained through an internet scraping process.
  • both the information for authentication and the image for comparison are each obtained through a user submission.
  • additional data relating to each of the user submissions may be obtained along with the information for authentication and the image for comparison, the additional data comprising identifying information about each user, a date and time of each user submission, and each user’s method of submission.
  • At least one of the marking sequence, the unique marked copy, the digital image, the information for authentication, the digital image for comparison, the authenticity conclusion report, and the additional data relating to each of the user submissions is stored in digital form in a database.
  • Some preferable embodiments of the present invention process PDF and/or other non-raster image files much faster by foregoing the character recognition and object grouping operations of the authentication process.
  • PDF files for example, often contain internal operators that group text and other objects, which the present invention can utilize to speed up its processing. Marking a PDF file may thus be completed without undertaking the optical character recognition and object grouping steps of the process. Instead, such non-raster image files provide the object grouping and the authentication process is streamlined to simply generate the sequence of modifications and persist it in the database, apply the modifications to the image file, and assemble the marked image file for use.
  • the marking results of the present invention is identical to those apparent on raster-format files.
  • the process utilizes the non-raster image file’s internal character recognition and object grouping operators, the authentication process can be sped up to ten times faster for non-raster image files than that applied to raster-format files.
  • this improved processing time can be vital when processing large amounts of data amongst one or more files.
  • the objectives of the present invention are further achieved by providing a system for authenticating digital information.
  • One preferable embodiment of such system includes a preliminary analysis module for obtaining and processing the digital information and converting it into a digital image, and an authentication marking module for isolating and grouping segments of content from the digital image and generating and applying a marking sequence to the digital image to create a unique marked copy.
  • the preliminary analysis module obtains the digital information and converts it into a digital image.
  • the authentication marking module identifies segments of content from the digital image and groups the segments of content into one or more segment groups, and generates the marking sequence to shift at least one of the one or more segment groups in one or more directions, creating the unique marked copy of the digital image.
  • the marking sequence may also alter a spacing between two or more of the segments in at least one of the one or more segment groups.
  • the system may further include a comparative analysis module for authenticating a digital image for comparison.
  • the comparative analysis module obtains the digital image for comparison, initiates an authentication process, which includes comparing the digital image for comparison with the unique marked copy of the digital image, and generates an authenticity conclusion report identifying the digital image for comparison as authentic if the authentication process determines that it matches the unique marked copy and identifying it as inauthentic if it does not.
  • the authentication process may include digitally overlaying the digital image for comparison with the unique marked copy.
  • the comparative analysis module may obtain the digital image for comparison by employing an internet scraping protocol.
  • the system may obtain both the information for authentication and the digital image for comparison from an independent user submission. Additional data relating to each of the independent user submissions may be obtained along with the information for authentication and the image for comparison, the additional data including identifying information about each user, the date and time of each user submission, and each user’s method of submission.
  • at least one database is also provided, wherein at least one of the marking sequence, the unique marked copy, the digital image, the information for authentication, the digital image for comparison, the authenticity conclusion report, and the additional data relating to each of the user submissions is stored in the at least one database.
  • the system is implemented on a user’s device using local installation software.
  • the system is implemented over a network connecting a user’s device with a remote server.
  • Figure 1 is a general diagram describing how to mark visual images and identify/authenticate.
  • Figure 2 is a sample document to demonstrate the method.
  • Figure 3 shows how the visual image is separated (split) into its constituent parts by the example of the document from Figure 2.
  • Figure 4 shows examples of combining parts into groups using the example document from Figure 2.
  • Figure 5 demonstrates how the labeled copies of the document from Figure 2 are distinguished through the coloring of copies (the red copy is superimposed on the blue copy).
  • Figure 6 is a schematic representation of a method according to the preferable embodiments of the present invention depicted in Figures 1-5.
  • Figure 7 is a schematic representation of an application of the method according to preferable embodiments of the present invention depicted in Figure 6.
  • Figure 8 is a schematic representation of a system according to preferable embodiments of the present invention depicted in Figures 1-7.
  • Figure 9 is a schematic representation of a system according to the preferable embodiments of the present invention depicted in Figures 1-8.
  • Figure 10 shows how the constituent parts of a non-raster image file, such as a PDF, are separately identified and combined into groups.
  • Figure 11 is a schematic representation of a method according to the preferable embodiments of the present invention depicted in Figures 1-5 and 10, with the shaded boxes representing operations that are unnecessary when processing a non-raster image file, such as a PDF.
  • Figure 12 is a schematic representation of a method according to the preferable embodiments of the present invention depicted in Figures 1-5 and 10, with the unnecessary operations for processing a non-raster image file, such as a PDF, removed.
  • Figure 13 is a schematic representation of a system according to preferable embodiments of the present invention depicted in Figures 1-12.
  • the given method includes the following stages:
  • Stage 1 Preliminary analysis (analysis and structuring) of the marked information (document, drawing, scheme, etc.)
  • Stage 2 Marking of the image and the formation of a unique copy (which has unique geometric characteristics of graphic objects), with the registration of marking parameters and information about the recipient of the image (subject) in the journal.
  • Stage 3 Digital image analysis of a fragment of a marked copy (for example: photograph of the image on the screen or in printed form, scanned copy of the image in printed form, screen shot (screenshot) of the computer) and comparison with the registered marked images with the identification of the corresponding journal log entry.
  • a fragment of a marked copy for example: photograph of the image on the screen or in printed form, scanned copy of the image in printed form, screen shot (screenshot) of the computer
  • the present method is carried out using the novel computer software "LeaksID".
  • the software is one of the method realizations, but there can be other ways for realization.
  • the software provides the ability to download (transfer to the program) the user information in digital form (image, document, drawing, diagram, presentation); users receive marked (uniquely) copies of previously downloaded information; download for analysis of a digital image of information in full form or fragment (photographs of a screen or a printed copy, scan-image of a printed copy, etc.); identification marking on the analyzed image and the identification of the corresponding journal log entry.
  • the method does not exclude and suggests the possibility of a different implementation, also in the form of one or more computer programs, within which the claimed marking method, but use cases will be different (for example: marking of images will occur in a hidden (implicit) way for the user, without his conscious participation, thereby achieving an additional level of security).
  • Stage 1 Preliminary information analysis.
  • the marked information in digital form is transmitted to the software for analysis and formation of the structure of its visualization.
  • the original unmarked information (the original) is stored in a protected storage area, in a format suitable for optimal marking speed (for example, a multi-page scan-image of a document may be divided into separate pages), but not necessarily.
  • Marked information means any information that can in one way or another be displayed both on a screen and displayed on a printing device (For example: raster or vector format image, document, diagram, chart, drawing, etc.)
  • Step 1 A visual representation of the original (visual image) in any form (bitmap, raster image, vector image, etc.) is formed in the computer memory. If the information in the visual display consists of several pages, each page is processed in the same way.
  • Step 2 The visual image is divided into independent related components. For example, for a regular text document, this could be letters, words, signs, lines, etc. In the framework of the given method, an independent component is considered to be, in a general sense, the area of a visual image that can be moved relative to other "unrelated" components without losing the meaning of information or obvious visual defects. Any method can be used to identify components (for example: OCR - Optic Character Recognition). Additional steps for a visual image processing can be performed during this step in order to increase quality of component definition (for example: image binarization performed by any suitable method).
  • Each component is provided with data for enabling subsequent restoration of its coordinates on a visual image (for example: for raster images, this may be a pair of coordinates (x1 , y1 ; x2, y2) of diagonally opposite angles of the framing rectangle, or coordinates (x, y) of one angle and characteristics of the width and height of the framing rectangle, etc.).
  • Step 3 Forming of the structure of groups of dependent components. During this Step, all components obtained in Step 2 are connected into dependent structures.
  • the dependent structure in the framework of the present method means a set of components that must be moved along a particular axis of the image in a single direction and a single distance (for example: all characters in one line of the document can only be moved vertically together; or the first words of all lines of a paragraph except the first, can be moved horizontally only together).
  • the grouping of dependent components can occur in various ways and methods (for example, for the black-and-white image of a document, you can use the graphical spectrum method described in O'Gorman, L. “The document spectrum for page layout analysis” (Volume: 15, Issue: 11) or any other suitable method. In such a way, as a result of this step, a structured representation of the geometric characteristics of the information visual image is obtained, which is stored in the computer memory and is associated with the original (loaded earlier in Step 1).
  • Marking of the image takes place in the background (preliminary) or upon request (either from a user and from other programs, in the case of embedding, or both).
  • the original visual image is modified by displacing groups and components along different axes, but preserving the dependencies (determined at the 3rd Step of the 1st Stage).
  • paragraph i For a word document: for paragraph i, increase the line spacing by 0.1 points, j the gap of paragraph k, increase by 50% in proportions, add s spaces before the t-th word in m paragraph, etc.
  • the present method does not limit methods of changing the geometric characteristics of the visual image.
  • the above examples are intended only to clarify the purpose and method of marking.
  • the amount of possible information is directly proportional to the number of blocks on the visual image of the original.
  • the sequence on the basis of which the original was marked (by displacing blocks and groups) is stored in a protected storage area and is uniquely associated with the transaction log, in which facts, time and subject that requested the marked image of information are recorded.
  • the program that implements the present method loads the analyzed visual image, i.e. “sample” (for example, a screen picture ora scan image created from a paper document).
  • sample for example, a screen picture or a scan image created from a paper document.
  • the original information previously downloaded into the program is then selected (the selection can be made by the user or the program itself on the basis of various methods, for example, using full-text search).
  • Comparison can take place in various ways, including by overlaying a sample on a previously reproduced copy, in which case the degree of similarity can be calculated by counting overlapping parts of the image (for example: the number of matched pixels for bitmaps).
  • sample and the recreated copy can be processed in various ways (for example, binarized).
  • Figure 1 depicts a schematic representation of the steps of an exemplary embodiment of the presently invented authentication process.
  • the original data or information 12 is obtained through digital means, preferably from a user uploading the original data or information 12.
  • the original data or information 12 may then be converted to a format depicting its content as a digital image 14, such as a pdf, jpeg, gif, tiff, or other similar file type, as will be known to those of ordinary skill in the art.
  • the present invention then generates one or more marking sequences 16 and applies them to the digital image 14, creating one or more unique marked copies 18 of the digital image.
  • the system preferably stores the original data or information 12, the unmarked digital image 14, the one or more marking sequences 16, and the one or more unique marked copies 18 in a database or similar digital data storage structure and preferably associates the stored information with identifying information about the user that submitted the original data or information 12 and the circumstances surrounding that submission, such as a date/time stamp, method of submission, etc.
  • the present invention may then compare a subsequently obtained digital image with the stored marking sequences 16 using an authentication process 20 to determine if the subsequently obtained digital image is an authentic copy and, preferably, identify the user that initiated the matching marking sequence 16 to determine if the subsequently obtained digital image is an authorized copy, if the source of the subsequently obtained digital image is associated with said user, etc.
  • Preferable embodiments of the authentication process 20 include overlaying the subsequently obtained digital image using the one or more unique marked copies 18.
  • the authentication process 20 may be employed using only a fragment of the subsequently obtained digital image or may compare the entire subsequently obtained digital image to only a fragment of the one or more unique marked copies 18.
  • Figure 2 depicts an exemplary digital image 14 demonstrating the application of the present technology.
  • the digital image 14 depicted in Figure 2 contains text, a table, and several shapes. Each of these objects in the digital image may be isolated, fragmented, and grouped in various ways in preparation for the authentication process of the present invention.
  • Figure 3 depicts the same exemplary digital image 14 as depicted in Figure 2. However, in Figure 3, each text object has been isolated and identified. The various text objects may now be grouped in various ways with each other and with the other table and shape objects in the digital image.
  • Figure 4 again depicts the same exemplary digital image 14 as depicted in Figures 2 and 3.
  • each of the table and shape objects have likewise been isolated and identified.
  • Figure 4 also depicts the grouping of the various objects, represented by the colored boxes surrounding each object.
  • the present invention can apply the marking sequence 16, shifting one or more groups in a predetermined direction, indistinguishable upon visible inspection of the resulting unique marked copy 18 of the digital image but plainly recognizable using the overlay procedure of the present invention’s authentication process 20.
  • Figure 5 depicts an exemplary unique marked copy 18 overlayed upon the exemplary digital image 14 depicted in Figures 2-4. As shown, the slight shift of certain groups of objects in the unique marked copy 18 become plainly visible upon performing the overlay comparison authentication process 20, although the same shift would be practically unrecognizable upon visual inspection of the unique marked copy on its own.
  • FIG. 6 an exemplary method according to preferable embodiments of the present invention is depicted schematically.
  • the process begins by obtaining a digital image 14 or by obtaining original data or information 12 in digital form and converting the original data or information 12 into a digital image 14 through the step of rendering the image 22.
  • the digital image 14 is analyzed to define image segments or fragments 24, and the segments or fragments are then organized into segment groups 26.
  • the present invention next generates a marking sequence of modifications 28 and applies the sequence to the image segment groups, modifying the image 30.
  • a unique marked copy 18 of the digital image 14 is created, rather than marking the original digital image 14.
  • Preferable embodiments of the present invention also store the digital image in a database 34 at various points in the process as well as the marking sequence for later reference.
  • FIG. 7 an exemplary application of the preferable embodiment of the method depicted schematically in Figure 6 is depicted, also schematically.
  • the original digital image 14 is first analyzed to identify image segments or fragments 36, depicted in dashed-line boxes. Those segments or fragments 36 are then organized into segment groups 38.
  • the segment groups 38 are defined line by line, again depicted in dashed-line boxes.
  • the segment groups 38 are altered according to the marking sequence, resulting in a unique marked image 18.
  • the segment groups 38 are shifted vertically relative to one another, and the spacing between the segments or fragments 36 within each segment group 38 has been altered.
  • Figure 8 depicts a schematic representation of an exemplary system 40 according to preferable embodiments of the present invention.
  • the exemplary system 40 depicted in Figure 8 can implement and execute the exemplary processes and methods depicted in Figures 1-7.
  • Preferable embodiments of system 40 include a preliminary analysis module 42, an authentication marking module 44, and a comparative analysis module 46.
  • the preliminary analysis module 42 obtains the original data or information 12 for processing. If the original data or information 12 is obtained from a user, preferable embodiments of the preliminary analysis module 42 also obtain information about the user and the circumstances surrounding the user’s submission and request, such as a date/time, method of submission, etc.
  • Those of ordinary skill in the art will recognize the various ways of obtaining such information from the user, who may also volunteer such information.
  • the preliminary analysis module 42 determines if the original data or information 12 is in an appropriate digital image 14 format. If not, the preliminary analysis module converts the original data or information 12 into an appropriate digital image format 14. The preliminary analysis module 42 then preferably stores the original data or information 12, the original digital image 14, and/or the information relating to the user and the user’s submission in a database 34 for later reference and creates a copy of the original digital image 14 for processing. [0098] The system 40 next analyzes the digital image 14 copy and identifies the various segments or fragments 36 found therein. This step may be performed by either the preliminary analysis module 42, the authentication marking module 44, or the two in combination.
  • the preliminary analysis module 42 may perform an optical character recognition (OCR) process on the digital image 14 copy, and the authentication marking module 44 may then isolate and identify the various segments or fragments 36 in the document. Or, either the preliminary analysis module 42 or the authentication marking module 44 may do both.
  • OCR optical character recognition
  • the authentication marking module 44 groups them into one or more segment groups 38. Some preferable embodiments may base such groupings on, for example, the nature of the individual segments 36 (e.g. text, tables, shapes, other objects), location within the document (e.g. segments 36 of text on a single line are grouped together, etc.), and/or mobility relative to other segments or fragments 36. Those of ordinary skill in the art will recognize the various approaches to creating the segment groups 38.
  • the authentication marking module 44 can then create the marking sequence 16 and apply it to the digital image 14 copy, creating a unique marked copy 18 of the original digital image 14. Both the marking sequence 16 and the unique marked copy 18 are preferably stored in the database 34. In the case of a user-initiated application of the system 40 in which data about the user and the user’s request was recorded and stored, the authentication marking module 44 preferably stores the marking sequence 16 and the unique marked copy 18 in association with said user data. The unique marked copy 18 is then also provided to the user who initiated the request for his or her use.
  • the system 40 preferably employs a comparative analysis module 46, though the steps performed by the comparative analysis module 46 could also or alternatively be performed by the authentication marking module 44.
  • the comparative analysis module 46 first obtains the image for comparison 48.
  • the image 48 may be obtained via user submission, using an internet scraping process, or through other electronic means, as will be understood to those of ordinary skill in the art.
  • the comparative analysis module 46 next preferably performs the authentication process 20. Preferable embodiments authenticate the image for comparison 48 by overlaying it with the one or more unique marked copies 18 previously created by the authentication marking module 44 and stored in the database 34. The comparative analysis module 46 may also perform the authentication process 20 using the original digital image 14 as well, where applicable. Upon identifying a unique marked copy 18 matching the image for comparison 48, preferable embodiments of the comparative analysis module 46 generate an authenticity conclusion report 50 confirming the image for comparison’s 48 authenticity. If no match exists, the authenticity conclusion report 50 identifies the image for comparison 48 as an unauthorized copy.
  • the authenticity conclusion report 50 is preferably provided to the user in the case of a user-initiated authentication process 20 and is also preferably stored in the database 34 for later reference.
  • the authenticity conclusion report 50 preferably includes information on the unique marked copies 18 used in the authentication process 20 and which, if any, matched the image for comparison 48.
  • the authenticity conclusion report 50 may also reflect any other data associated with any matching unique marked copy 18, such as the user data and user request data associated with such unique marked copy 18, where applicable.
  • the system 40 may take further action to attempt to remove the unauthorized image 48, such as initiating a take-down request on a third-party website, taking steps to initiate legal proceedings, etc.
  • the steps that may be automated to protect the authenticity of the digital images 14 submitted to the system 40 for protection.
  • the system may be implemented locally using local installation software or implemented over a network connecting a local device with a remote server.
  • a user device 62 can be connected to the system 60 over a network 64, although the system may also be installed locally on the user device 62.
  • the system 60 preferably has a gateway 66 and a frontend service 68 providing the user device 62 access to the system 60.
  • a user may access the system 60 using a web-browser through a web interface generated by the front-end service 68.
  • the gateway 66 and/or frontend service 68 may additionally authenticate user credentials and perform other access security functions, as will be understood by those of ordinary skill in the art.
  • the gateway 66 may provide user devices 62 with secured access to the system 60 using HTTP(S) protocol and operate JSON (JavaScript Object Notation) format data.
  • JSON JavaScript Object Notation
  • the exemplary system 60 preferably employs a preprocessing service 70 that can identify and convert, if necessary, any information provided by the user into a digital image file.
  • the preprocessing service 70 may also perform other document processing functions, such as OCR (optical character recognition), etc., and/or a document service 72 may perform such document processing functions.
  • the document service 72 also preferably implements functions to manage the other services of the system and preferably manages registration and storage of information about the digital image file using a service registry 74, a file service 76, and a file storage 78.
  • the file service 76 also preferably provides unified file storage functionality for the system 60, and the service registry 74 preferably implements “service discovery” functionality.
  • a marking service 80 next processes the digital image provided by the preprocessing service 70, identifying the various segments and/or segment groups found in the digital image file.
  • the marking service 80 then marks the digital image file, as discussed above, creating the unique marked copy, and preferably stores the unique marked copy and the marking sequence in a database 82.
  • a configuration service 84 may assist in developing the marking sequence and preferably provides configuration parameters to the other services employed by the system 60.
  • An analysis service 86 operates to authenticate an item submitted from the user device 62.
  • the analysis service 86 preferably queries database 82 to obtain digital image data for comparison and performs the authentication process, preferably by overlaying the item submitted for authentication with the digital image data obtained from the database 82.
  • the system’s front-end service 68 preferably also provides a message service 88 permitting the system’s 60 components to interact with each other in an asynchronous way and preferably also facilitates interaction with the user devices 62 connected via the network 64.
  • Such message service 88 may, for example, provide the authenticity conclusion report created by the analysis service 86 to a user seeking an authenticity confirmation on the item submitted for authentication.
  • the components of system 60 may perform these and other functions to implement the full extent of functionality of the invented method and system.
  • the present invention may process certain non-raster format digital documents, such as PDFs, MS Word files, MS PowerPoint files, etc., somewhat differently from the processing of raster-format digital documents in some preferable embodiments.
  • PDF files are often embedded with character recognition and object grouping functions and operators, which allow the present invention’s authentication system and process to mark such files without undertaking the character recognition and object grouping steps, speeding up the authentication process substantially.
  • the marking of non-raster image files and raster files creates the same output.
  • the authentication process can be up to ten times faster for non-raster image files, which is highly desirable particularly when processing large amounts of data, as will be recognized by those of skill in the art.
  • FIG. 10 depicts an exemplary portion of a digital file, such as a PDF.
  • the exemplary PDF file format contains groups of objects.
  • the PDF file format is a sequential set of operators, and PDF viewers process these operators sequentially such that PDF files appear identical whatever device is used to view them.
  • all visual objects are grouped through CM operators (transformation matrix that sets the rules for displaying and scaling nested objects).
  • CM operators transformation matrix that sets the rules for displaying and scaling nested objects.
  • the objects (expressed through a set of different operators) in the PDF are divided into groups through the operators cm, Q and q, and the movement of an entire group of objects can be accomplished by changing the parameters of the operator cm.
  • the framing operators BT and ET are used within which new output parameters can be set, through the operators Tm (text transformation matrix).
  • Tm text transformation matrix
  • all operators between the encountered operators Tm also form a group and can be labeled by changing the parameters of the operator Tm.
  • all the text is output through the operators Tj and TJ, which output specific text lines. Text lines represent nested groups and are preferably marked using the present invention’s method by moving the drawing start point through the Td statement.
  • Tj operator lineoutput operator
  • TJ operator array-of-string operator
  • each of these operations is used in preferable embodiments of the present invention to mark PDFs and other applicable non-raster image files.
  • the present invention can omit the steps of the authentication process associated with rendering an image, identifying the individual elements therein, and segmenting and grouping of those individual elements.
  • preferable embodiments of the present invention can skip directly to generating the sequence of modifications and applying those modifications to the non-raster image file at the file level, substantially improving processing speeds for large amounts of data.
  • FIG. 11 depicts an exemplary and schematic representation of a method of the present invention as applied to a non-raster image file, such as a PDF.
  • a non-raster image file such as a PDF.
  • the shaded boxes relating to rendering the image 22, defining image segments 24, defining segment groups 26, and persisting the segments and groups in the database 34 are not required for non-raster image files. Instead, the present invention uses the groupings already present in the native PDF or other non-raster image file to define the groups and prepare the file for marking. As depicted in FIG.
  • the authentication process instead receives the digital PDF or other non-raster image file 14, extracts every operator 25 to define groups for marking using the non-raster image file’s 14 inherent properties and operators, generates the sequence of modifications 28 and applies those modifications for each operator 30, and reassembles the operators to the file 32, creating the marked PDF or other non-raster image file 18.
  • the process also preferably persists the modification sequence in the database 34 for future reference.
  • Processing time and resource efficiency is further improved by omitting the step of persisting the segments and groups in the database 34 as well.
  • the same comparison operation is employed but without the added burden of accessing and processing the segmentation and grouping steps applied and persisted in the database (22, 24, 26, 34) as is necessary for rasterformat files.
  • FIG. 13 depicts a schematic representation of an exemplary system 40 according to preferable embodiments of the present invention capable of processing PDF and other non-raster image files at speeds upwards of ten times faster that required for processing raster-format files.
  • the preliminary analysis module 42 may be unnecessary in some preferable embodiments when processing only PDF or other non-raster image files.
  • System 40 preferably operates to receive the digital image file 14, and either the preliminary analysis module 42 or, in some preferable embodiments, the authentication sequence module 44 groups the objects and text using the operators from the non-raster image file. The authentication sequence module 44 then preferably generates the marking sequence 16, persists the marking sequence 16 in the database 34, and applies the marking sequence 16 to the digital image and reassembles the operators to create the marked copy 18 of the PDF or other non-raster image file.
  • the system 40 preferably employs a comparative analysis module 46 to assess the image’s 48 appearance based upon the marking sequences 16 stored in the database 34, performing the authentication process 20 and generating an authenticity conclusion report 50. If the image for comparison 48 matches one of the saved marking sequences 16 from the database 34, the system 40 generates an authenticity conclusion report 50 determining that the file is authentic. If no match is found, the authenticity conclusion report 50 would identify the image for comparison 48 as an unauthorized copy. This entire process is performed far more rapidly with respect to PDF and other non-raster image files by omitting the steps needed for pre-processing raster-format files, including rendering the image 22, defining image segments 24, defining segment groups 26, and persisting the segments and groups in the database 34.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Editing Of Facsimile Originals (AREA)

Abstract

La présente invention porte sur un procédé d'authentification d'informations numériques qui consiste à : obtenir, sous forme numérique, des informations en vue d'une authentification; préparer les informations en vue d'un traitement, une telle préparation comprenant la conversion des informations en une image numérique; identifier des segments de contenu dans l'image numérique; regrouper les segments de contenu en un ou plusieurs groupes de segments; générer une séquence de marquage comprenant le déplacement d'au moins un groupe parmi les un ou plusieurs groupes de segments dans une ou plusieurs directions; et appliquer la séquence de marquage à l'image numérique, créant ainsi une copie marquée unique de l'image numérique.
PCT/US2022/075001 2021-08-16 2022-08-16 Procédé de marquage d'éléments visuels d'informations en vue d'une identification ou d'une authentification ultérieure Ceased WO2023023502A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US17/403,127 US11606475B2 (en) 2019-06-06 2021-08-16 Method for marking visuals of information for subsequent identification or authentication
US17/403,127 2021-08-16

Publications (1)

Publication Number Publication Date
WO2023023502A1 true WO2023023502A1 (fr) 2023-02-23

Family

ID=85239794

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2022/075001 Ceased WO2023023502A1 (fr) 2021-08-16 2022-08-16 Procédé de marquage d'éléments visuels d'informations en vue d'une identification ou d'une authentification ultérieure

Country Status (1)

Country Link
WO (1) WO2023023502A1 (fr)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170091783A1 (en) * 2015-09-24 2017-03-30 Authenticate, Llc. Three-Dimensional Authentication and Identification Methods, Devices, and Systems
US20180186167A1 (en) * 2016-12-30 2018-07-05 Morphotrust Usa, Llc Embedded variable line patterns
US20200387999A1 (en) * 2019-06-06 2020-12-10 G-71 Inc. Method For Marking Visuals Of Information For Subsequent Identification Or Authentication

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170091783A1 (en) * 2015-09-24 2017-03-30 Authenticate, Llc. Three-Dimensional Authentication and Identification Methods, Devices, and Systems
US20180186167A1 (en) * 2016-12-30 2018-07-05 Morphotrust Usa, Llc Embedded variable line patterns
US20200387999A1 (en) * 2019-06-06 2020-12-10 G-71 Inc. Method For Marking Visuals Of Information For Subsequent Identification Or Authentication

Similar Documents

Publication Publication Date Title
EP3673391B1 (fr) Protection du droit d'auteur basée sur des informations de droit d'auteur masquées
US7639388B2 (en) Image processing apparatus, image reproduction apparatus, system, method and storage medium for image processing and image reproduction
US8014560B2 (en) Preserving scanner signature using MRC technology
US10692167B2 (en) System and method for digitally watermarking digital facial portraits
US9864920B2 (en) Semiautomatic multifunction device-based validation of secure documents
CN111160860A (zh) 一种文件的打印及盖章方法
US9652814B2 (en) Method and apparatus for generating and authenticating security documents
CN115114598A (zh) 水印生成、水印用于文件追溯的方法及装置
CN110321881B (zh) 识别包含身份证明证件的图像的系统和方法
US11120520B2 (en) Method for marking visuals of information for subsequent identification or authentication
US6631200B1 (en) True-life electronics signatures
CN111339556A (zh) 数据脱敏方法、终端、装置及存储介质
US11606475B2 (en) Method for marking visuals of information for subsequent identification or authentication
US10460163B2 (en) System and method for digitally watermarking digital facial portraits
CN114399278A (zh) 一种工程图纸的图框智能定位及批量签章方法
WO2023023502A1 (fr) Procédé de marquage d'éléments visuels d'informations en vue d'une identification ou d'une authentification ultérieure
JP7392322B2 (ja) 画像処理装置、画像処理方法、画像処理装置用プログラム、および、書類管理システム
WO2008154381A1 (fr) Mécanisme de sécurisation de données encodées
KR100697879B1 (ko) 픽셀을 이용한 데이터의 암호화 및 해독방법
TWM634167U (zh) 文件比對系統
EP0980179A1 (fr) Procédé d'adjonction d'information dans une image et procédé d'extraction d'information d'une image
JP2017097560A (ja) 情報処理装置、情報処理システム、情報処理方法及び情報処理プログラム
US20250286723A1 (en) System and methods for detecting altered documents
JP4260076B2 (ja) 文書作成装置、文書検証装置、文書作成方法、文書検証方法、文書作成プログラム、文書検証プログラム、文書作成プログラムを格納した記録媒体、および文書検証プログラムを格納した記録媒体
US20250285461A1 (en) System and methods for detecting altered documents

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22859312

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22859312

Country of ref document: EP

Kind code of ref document: A1