KR101835105B1 - Method for identification of counterfeit print matter - Google Patents

Abstract

In order to prevent forgery of an image printed with a halftone dot printed on an image obtained by scrambling a plurality of divided images of an original image, a position of a divided image for recognition and a photographed image of a printed matter is registered in advance, The present invention relates to a method of identifying a false imprint based on a divided image pattern or a halftone dot at a position registered for recognition.

Description

METHOD FOR IDENTIFICATION OF COUNTERFEIT PRINT MATTER

In order to prevent forgery of a printed image obtained by scrambling a plurality of divided images of an original image, a divided image for identification is specified for each printed matter, and based on a pattern or a dot of a specific divided image, And to a method for identifying a forged print.

In recent years, we have been using printed materials for counterfeit prevention in sales products, publications, currency, securities, identification cards, and cards. However, as technologies related to scanners, printers, and design software are advanced, counterfeit prints This illegal use is constantly occurring.

However, the counterfeit printed matter can be minimized only if it is easily identified on the spot, and thus, a smartphone-based detection technology is disclosed.

Open Patent No. 10-2015-0077592 and Japanese Patent Application Laid-Open No. 10-2014-0072204 disclose a method of acquiring a photographed image of a printed matter with a mobile terminal, extracting characteristics of various security elements from the acquired photographed image, Judgment is made according to the result of comparison with element feature points.

However, the disclosures of Japanese Patent Application Laid-Open Nos. 10-2015-0077592 and 10-2014-0072204 disclose that the accuracy of identification, the provision of rapid identification results, ease of use in various applications, accurate identification of counterfeit prints by simple methods, There is a lack of satisfying all the requirements for actual implementation using smartphone technology, such as providing step-by-step results of smartphone performance.

KR 10-2015-0077592 A 2015.07.08. KR 10-2014-0072204 A 2014.06.12.

Accordingly, a problem to be solved by the present invention is to provide a method of providing a fast identification result, providing convenience of use, employing a simple method, and providing a stepwise result of recognition according to a constraint while easily recognizing a counterfeit printed matter without error, The present invention provides a method for identifying a counterfeit printed matter.

According to another aspect of the present invention, there is provided a method of predicting a false imprint on a genuine printed matter in which an original image is divided into a plurality of divided images and a puzzle image, (S100) for registering and storing, at a request of the registrant terminal, a contrast position for designating a divided image to be used for judging authenticity among the plurality of divided images of the puzzle image, which is a photographed image of the original print, ; A step S200 of receiving a sensed image as a sensed image of a printed matter to be authenticated having a standard of the genuine printed matter, from a server according to a sensory request of a user terminal; A genuine photographed image matching the identification code of the photographed image to be photographed is selected from the genuine photographed images registered in the server, and a divided image of the contrast position is extracted for each of the photographed image and the genuine photographed image, (S300) of judging whether or not the photographed object image is true or false and notifying the judgment result to the user terminal.

The pattern represented by at least the divided image of the contrast position is designed to extend to an adjacent region of the original image, and the determining step (S300) further extracts a divided image adjacent to the divided image of the contrast position on the original image And the pattern continuity between the post-split images is applied to the determination of authenticity.

The pattern comparison in the determining step (S300) includes a line thickness comparison.

In the original registration step S100, the plurality of contrast positions are designated, and the divided images of the contrast positions are adjacent to each other in the original image, and the determining step S300 is a pattern comparison for each of the divided images of the plurality of contrast positions And the pattern continuity between the divided images is also compared to determine the authenticity of the object image to be sensed.

In the determining step S300, a halftone dot comparison is performed after pattern comparison of the divided images at the contrast position to determine whether the sensed object image is authentic or not. The halftone dot comparison is performed based on the halftone dot spacing, the number of halftone dots, At least one of them is compared.

The original printed matter is a halftone dot printed after a specific color is further applied to at least one boundary line in the division boundary line of the puzzled image, and the position of the divided boundary line on which the specific color and the specific color are applied is registered in the server (S100) The recognition step S300 identifies and extracts the divided image at the contrast position using the color-based boundary line based on the registered color and the position of the dividing boundary line, and uses the reference line as the reference line, and compares the divided images .

As described above, according to the present invention, only the divided images at the contrast positions are precisely checked. Therefore, it is possible to quickly and accurately identify the errors, thereby eliminating the error of the recognition. Also, by printing a puzzled image, In particular, since the contrast position is not exposed and the halftone dot at the boundary of the divided image at the contrast position is deformed as compared with the image before printing, precise copying or duplication of the divided image can be prevented in principle .

Further, it is difficult to restore the puzzle text of the printed matter into the original image form, and the continuity of the pattern is applied to the recognition, so that accurate recognition results can be obtained.

In addition, the present invention provides a result of the pattern matching and the comparison of the halftone dots in a stepwise manner, and provides a result of the test in accordance with the resolution of the captured image obtained by the user terminal, so that the result of the test can be obtained even by a user terminal of various types.

Further, according to the present invention, even when the image is copied by photocopying, photographed, photographed, photographed or photographed image, it is possible to accurately compare the fogged prints with the halftone dot spacing, the number of halftone dots, can do.

1 is a block diagram of a server for carrying out the present invention;
2 is a design image 10;
FIG. 3 is a halftone printed print 11 after puzzling the design image 10 of FIG.
4 is a flowchart of a method of detecting a forged print according to an embodiment of the present invention.
FIG. 5 is a detailed flowchart of the step S300.
6 is a genuine restoration image restoring a puzzle fingerprint area of a genuine photographed image obtained by photographing the original printed matter 11;
7 is an example of a forgery-subject printed matter 21 to be inspected.
FIG. 8 is a comparative image showing a partially enlarged portion of the printed material 11 of FIG. 3 and the counterfeit printed object 21 of FIG.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

FIG. 1 is a block diagram of a server 300 for carrying out a method of detecting a fake print according to an embodiment of the present invention.

Referring to FIG. 1, a false imprint detection method according to an embodiment of the present invention registers a genuine photographed image, which is a photographed image of a genuine printed matter 11, at a request of a registrant terminal 100 (S100) When the subject image to be inspected which is a photographed image of the printed matter 21 is received from the user terminal 200 through the communication network 400 and received a diagnosis request (S200), the genuine photographed image registered by the registrant terminal 100 The server 300 compares the photographed object image and determines whether the image is authentic (S300).

First, the shape of the original printed matter 11 and the grounds for utilizing it as counterfeit prevention will be described.

Fig. 2 shows an example of a design image 10 to be printed with the authentic original print 11 in the present invention.

Fig. 3 is a printed original 11 obtained by pounding the design image 10 of Fig. 2 and then performing halftone printing.

The design image 10 shown in Fig. 2 is a design image designed to obtain the original printed material 11, and can be designed with graphic software such as an illustrator. The original image 10a, 10a and the identification code image 10b pasted on the rim of the image 10a.

It should be noted that the original image 10a is an image having a predetermined pattern, and as shown in FIG. The color code related data is determined by designing the original image 10a on the PC.

A contrast position 10c is specified in the original image 10a. This contrast position 10c is specified to specify an area used for determining authenticity, as described later.

The identification code image 10b includes an identification code given to the original image 10a. Here, the identification code may be the product code of the product on which the original printed matter 11 is printed.

Meanwhile, according to the embodiment of the present invention, the identification code image 10b includes an IP address which allows the user terminal 200 to connect to the server 300 via the Internet of the communication network 400. [

According to the illustrated embodiment, the identification code image 10b is a QR code or a data matrix, and an identification code and an IP address are put on two sides having a position detection pattern, and a quadrangle-shaped data area The original image 10a can be placed in the design image 10 to complete the design image 10. Although not shown, the alignment pattern of the QR code can be assigned to a partial area on the original image.

In this design image 10, the original image 10a is divided into a plurality of divided images having the same rectangular shape by dividing the original image 10a into a lattice shape. Thereafter, the divided images are swapped, Image. Here, the plurality of divided images may not have the same shape or may not be rectangular, but an embodiment in which the same rectangular shape is used for the convenience of manufacturing will be described.

Next, a design image obtained by converting the original image 10a into a puzzled image is printed on the surface of the product as a halftone dot, and is shaped into the original printed matter 11 of Fig. 3, thereby authenticating the authenticity of the product to the original printed matter 11 Let's use it.

3, the identification code 11b, in which the identification code image 10b is printed and shaped, is in the form of an identification code image 10b, but after the original image 10a has been puzzled The printed and shaped puzzle copy 11a can not know the pattern of the original image 10a.

The contrast position 10c on the original image 10a is changed to the contrast position 11c displayed on the puzzle volume 11a of Fig. 3 by the puzzling process. This contrast position 11c becomes the position of the divided image used to judge whether or not the printed matter is authentic.

However, in order to identify the printed matter, it is necessary to accurately extract the divided image of the contrast position 11c.

Thus, the original printed matter 11 may be halftone printed after a specific color is further applied to at least one of the divisional boundary lines in the divisional boundary line of the puzzled image. At this time, it is preferable not to select the divided boundary line with a specific color as the boundary line of the divided image of the contrast position 11c.

The genuine printed matter 11 thus shaped can be used as an anti-falsification printing material as follows.

First, it is not possible to obtain the same illustration as the puzzle book 11a even if it is tried to scan the original printed matter 11 for the purpose of forgery or to obtain an illustration with the photographed image. This is because the original image 10a is not directly printed but the puzzle and divided image inversion are performed and then printed to obtain the puzzle copy 11a, so that the pattern is split into a discontinuous and difficult to recognize partial pattern.

Secondly, since the printed matter is identified using only the divided image of the contrast position 11c, the complexity or the amount of computation of the signal processing for the recognition can be reduced, and accurate detection results can be obtained by precisely comparing only the divided images. Also, by making the contrast position 11c invisible, it is also possible to fundamentally prevent falsification of only the divided image of the contrast position 11c.

Third, it prevents the same forgery of patterns and halftone dots according to the halftone printing.

According to the halftone printing, the skirts of the halftone dots, the size and the angle of the halftone dots are determined according to the size and resolution of the printed material, and the halftone dots called ink pockets are designed to be arranged in a lattice pattern. Then, a halftone dot printing is performed with a printing plate obtained by plate printing. However, it is known that the interval between the halftone dots in the plate making process can not be the same every time the plate is formed.

That is to say, even if the same original printed material 11 can not be obtained even by the same plate making process, the same design image 10 can not be obtained even if it is printed or printed using a scanned image or a shot image of the puzzle print 11a A counterfeit print having a halftone dot spacing can not be obtained.

In addition, since the halftone image is printed on a puzzled image, the halftone image is more difficult to duplicate, and it is not possible to place the halftone dot at the dividing boundary unless the halftone dot is formed on the boundary of the split image. Thereby making replication more difficult.

Thus, the divided image of the contrast position in the counterfeit print does not have the pattern of the original print, and when enlarged and compared, the halftone dot spacing, the position, and the count do not have the halftone dot of the original printed matter.

Also, it is impossible to reproduce the same hue of the hue that is used when printing with a printing plate obtained by plate printing, strictly speaking.

The present invention uses such a print for anti-falsification.

Hereinafter, an embodiment of the present invention using the above-mentioned original printed matter 11 will be described.

1, the server 300 for carrying out the present invention receives information on the original printed matter 11 in response to a request from the database unit 320 and the registrant terminal 100, and stores the received information in the database unit 320 The registration unit 310 for registration and management and the terminal 200 are connected via the Internet network of the communication network 400 to accept identification of the printed matter to be inspected 21 and then transmit the authentic printed matter 11 And a determinator 330 for notifying the discriminator terminal 200 of the result of the discrimination based on the information on the discriminator.

The registrant terminal 100 accesses the server 300 through the Internet network and transmits information on the original printed matter 11 to the server 300. The registrant terminal 100 is a terminal for requesting registration, .

The information on the original printed matter 11 includes a genuine shot image, a contrast position 11c, and a puzzle picture restoration rule.

The user terminal 200 may be a smartphone of a user who intends to identify a printed matter. Preferably, the object to be inspected is photographed to acquire an image of the object to be inspected, and then the image of the identification code 11b is extracted and decoded on the object image to be inspected to obtain the IP address of the server 300, An application for accessing the server 300 and requesting identification is installed in the user terminal 200. [ Such an application may be, for example, a QR code or a data matrix recognition application.

The detecting unit 330 of the server 300 detects a true photographed image corresponding to the identification code of the photographed image to be photographed in the genuine photographed image pre-registered in the database unit 320 when there is a request for identification of the user terminal 200 A divided image extracting unit 332 for extracting a divided image of the contrast position with respect to the object image to be sensed and the original image to be sensed, And an authenticity judgment unit 333 for notifying the judgment result to the smart phone.

A method of detecting a false imprint by the server 300 configured as described above will be described in detail with reference to FIG. 4 to FIG.

4 is a flowchart of a method of detecting a fake print according to an embodiment of the present invention.

FIG. 5 is a detailed flowchart of the determining step S300.

Fig. 6 is a restored original image restoring a picnic area of a genuine photographed image obtained by photographing the original print 11. Fig.

Fig. 7 is an example of a forgery-proof printed matter 21 to be inspected.

FIG. 8 is a partially enlarged view of a portion having contrast positions 11c and 12c with respect to the original printed material 11 of FIG. 3 and the forgery-tested printed matter 21 of FIG.

The method for identifying a forged article according to an embodiment of the present invention includes a step of registering a trial (S100) performed at the request of the registrant terminal (100), a step of requesting recognition S200 according to a request of the user terminal 200, (Step S300) of notifying the result of the identification of the printed matter.

The genuine registration step S100 is a step in which the registrant terminal 100 makes a request to register in the web environment the information about the original printed material 11 in the registering unit 310 when the registrant terminal 100 accesses through the Internet, In the database unit 320. [0050]

Here, the information on the original printed matter 11 is assigned to the puzzle copy 11a of the original print image 11 obtained by photographing the original printed matter 11, A contrast position 11c for designating a split image used for determination, and a puzzle picture restoration rule, which is a rule for restoring an image of the puzzle picture book 11a to the original image 10a.

Here, the pattern represented in the divided image of at least the contrast position 11c is not limited to be displayed only in the divided image, but rather is located at the contrast position 11c so as to extend to the area of the adjacent divided image on the original image 10a. . This is to make it difficult to copy and forge the divided image of the contrast position 11c.

Although only one of the contrast positions 11c is shown in the example figures of FIGS. 2 and 3, it is also possible to designate the plurality of contrast positions 11c.

When a plurality of the contrast positions 11c are designated, the divided images of the contrast positions 11c are made adjacent to each other in the original image. That is, when the image of the puzzle finger 11a is restored to the original image 10a, the divided images of the specified plurality of contrast positions 11c are adjacent to each other.

The puzzle thumbnail restoration rule divides the original image 10a shown in FIG. 2 into a plurality of divided images, and thereafter rearranges the plurality of divided images to generate a puzzle image shown in FIG. 2, And the reverse order of the inversion rule. That is, the original image 10a can be obtained by dividing and separating the divided images into puzzle-thumbnail images according to the puzzle thumbnail restoration rule, and rearranging the separated divided images in the reverse order of the inversion rule.

Of course, it is preferable to restore the puzzle fingerprint restoration rule to the original image including the size of the divided image and the dividing boundary line precisely.

On the other hand, when at least one of the divisional boundaries of the paginated image of the original printed material 11 is further colored with a specific color and then halftone printed, the specific color and the divisional border positions on which the color is applied are also printed on the original printed matter 11 ) To be included in the information about the information.

The registering unit 310 of the server 300 accepts the registration request every time a registration request is received by the registrant terminal 100 and stores the information on the original printed matter 11 in the database unit 320, Decodes the identification code image and stores it so that it can be retrieved by the identification code.

By registering the information about the original printed matter 11 in this way, the recognition requesting step 200 and the identifying step S300 can be performed.

The recognition request step 200 receives the image of the object to be inspected, which is a captured image of the object to be inspected, having the standard of the original print, and transmits the image of the object to be inspected to the user terminal 200 through the Internet network to be.

According to the embodiment of the present invention, since the original printed matter 11 conforms to the format of the QR code, if the user terminal 200 is a smart phone equipped with an application capable of recognizing the QR code, It is possible to easily perform the connection with the server 300 using the IP address of the identification code 11b and the transfer of the image of the object to be inspected.

The detecting step S300 is a step performed by the detecting unit 330 when receiving the sensing target image from the user terminal 200. The detecting step S300 is a step of detecting the distance between the sensing target image and the original sensing image, Judges whether the photographed object image is true or not, and notifies the judgment result to the user terminal.

5 is a detailed flowchart of the determining step S300.

First, the genuine sorting unit 331 decodes the identification code in the sensing target image (S310) and selects a genuine photographed image corresponding to the decoded identification code among the genuine photographed images stored in the database unit 320. [

Next, the divided image extracting unit 332 extracts the divided image of the contrast position 11c specified for the original photographed image from the original photographed image and the photographed target photographed image, respectively.

Of course, the divided image of the contrast position 11c is extracted by equally applying the puzzle fingerprint restoration rule and the contrast position 11c registered and stored corresponding to the original photographed image to the noticed original image as well as the original photographed image. That is, the contrast position 21c of the subject image to be sensed is the same as the contrast position 11c registered corresponding to the original photographed image.

If the position of the divisional border on which the specific color corresponding to the original photographed image and the specific color are registered is registered, the division boundary line of the divisional image in the contrast position 11c is more accurately identified based on the specific color and the divisional border line The split image can be extracted. That is, since the divided images are divided into the lattice form of equal intervals in the vertical and horizontal directions, the dividing boundary of the divided image in the contrast position 11c can be accurately identified based on the divided boundary line in which the specific color is applied. If such a partitioned border with the specific color is registered as the border of the divided image, it is preferable to remove the specific color from the halftone dot.

In addition, if only one contrast position 11c is registered, a divided image partitioning method can be applied during the puzzle picture restoration rule to extract a divided image of the contrast position 11c.

If the contrast position 11c is registered as two or more, the divided image partitioning method is applied during the puzzle picture restoration rules to extract the divided image of the position 11c at each contrast. Then, And rearranges them according to the reverse order of the puzzle inversion rule during the thumbnail restoration rule and restores to the position on the original image. As described above, in the embodiment of the present invention, a plurality of divided images that have been extracted and rearranged are recognized by each other.

Next, pattern comparison is performed between the contrast-divided image of the extracted original printed material 11 and the contrast-divided image of the object 12 to be inspected (S330).

The pattern comparison preferably includes a comparison of line widths appearing in the image. As described above, even if a halftone dot print is copied and falsified, it is practically impossible to forge the halftone dot spacing, the halftone dot position, the halftone dot count per unit area, and the color code value of the halftone dot equally. Because.

In the case of extracting and comparing divided images at a plurality of contrast positions, the divided images at the respective contrast positions are compared with each other, and at the same time, they are restored to positions on the original image and are adjacent to each other. do.

If the patterns are not matched as a result of the pattern comparison, the forgery determination (S332) is performed and the process proceeds to the following determination result generation step (S350). If the patterns match, the next pattern comparison steps (S340, S341, S342, S343, S344, and S345) I go over.

Here, the match is made when the difference is within the allowable value set in advance. That is, since the image quality of the object image to be detected varies according to the camera performance of the user terminal 200, the allowable value of the difference is preset in consideration of the camera performance of various types of the user terminal 200.

The results of pattern comparison will be described by way of example.

6 is a genuine reconstruction image (12) obtained by restoring the entire puzzle transform region of the original photographed image according to the puzzle reconstruction rule.

Referring to FIG. 6, it can be seen that the boundary image of the divided image is damaged due to restoration from the shot image of the halftone dot print, but it is almost the same as the original image 10a of the design image 10 shown in FIG.

In actual practice of the present invention, only the divided image of the contrast position 12c can be extracted. However, when the entire puzzle picture is restored according to the puzzle picture restoration rule registered in the server, .

It is also possible to judge whether the image is true or false by comparing the pattern continuity between the divided images extracted and extracted together with the divided image adjacent to the divided image of the contrast position 12c on the original restored image 12 of Fig.

Fig. 7 shows the image of the object to be inspected when the printed matter 21 to be inspected is a forgery-printed object.

FIG. 8 is an enlarged view showing a portion having a contrast position with respect to the photographed image of FIG. 7 and the genuine photographed image of FIG.

As shown in Fig. 7, the image of the subject to be inspected obtained by photographing the forged copy of the original printed matter 11 has a size of the original printed matter 11, that is, the size of the puzzle text 11a and the identification code 11b And has a puzzle book 21a and an identification code 21b in accordance with a specification and layout standard. Of course, it is preferable to verify whether the standard of the photographed image to be inspected conforms to the standard of the original print 11 before performing the determination step S300, and to perform the recognition step S300 only when the specification is the same.

As shown in Fig. 8, even if the print target is a forgery-proof object to be inspected, a pattern of sharp streaks is obtained in the contrast position 11c divided image obtained from the original photographed image. However, , It is possible to judge whether the printed matter to be inspected is authentic or not by comparing patterns.

The original printed matter illustrated in the drawing is 1.5 cm in width and 1.5 cm in height, and the puzzle pieces are divided into 12 puzzle pieces each in the horizontal and vertical directions and puzzled. In this case, in order to be able to compare the pattern with one divided image, it is necessary to have more than the number of halftone dots in the divided image to show the pattern in the divided image.

In the halftone comparison steps S340, S341, S342, S343, S344, and S345, the halftone dots are clearly identified in the divided image at the contrast position extracted from the image of the object to be identified, and it is determined whether the halftone dots can be compared (S340) (S341). If it is determined that the halftone dot comparison is possible, the halftone dot comparison is performed on the divided image at the contrast position extracted from the original photographed image and the split image extracted from the sensed target image (S342).

The halftone comparison is to compare at least one of the halftone dot spacing, the number of halftone dots, the halftone dot position, and the halftone dot color code value in the divided image.

The result of the comparison of the halftone dots (S343). If the halftone dots do not match, the counterfeit determination is made (S344). If they match, the true judgment is made (S345).

Whether or not a match to the halftone dot is determined is determined in advance according to the difference in camera performance of the user terminal 200, and it is determined that the tolerance matches if the difference of the halftone dot is within the allowable value.

Referring to FIG. 8 showing the results of the comparison of the halftone dots, the comparison of the halftone dots of the contrast positions 11a and 21c shows that there is a difference in the halftone dot spacing, the number of halftone dots, the halftone dot position, and the halftone dot color. This is because the difference of the halftone dots appears in the process of printing the photographed image of the original printed matter 11 or in the process of forging and printing the photographed image.

In step S350, the result of the pattern comparison and the comparison of the halftone dots is generated in step S350, and the result of the determination is transmitted to the user terminal 200 in step S360.

As described above, the result of the test is a true determination in the case where the patterns and the halftone dots are all matched, a case in which the patterns do not match, or a fake judgment in the case where the patterns match but the dots do not match, Pattern match / halftone dot comparison hold determination in the case of performing the comparison.

However, the pattern match / halftone comparison hold determination is because the halftone dot is not clear in the image to be sensed, so it is advisable to perform a guidance for requesting a sensed image with high image quality when notifying the recognition result.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, . ≪ / RTI > Accordingly, such modifications are deemed to be within the scope of the present invention, and the scope of the present invention should be determined by the following claims.

10: Design image
10a: Original image 10b: Identification code image 10c: Contrast position
11: Original printed matter
11a: Puzzle booklet 11b: Identification code 11c: Contrast location
12: Original restored image
12c: Contrast location
21: Printed object
21a: puzzle book 21b: identification code 21c: contrast position
100: Registrant terminal
200: User terminal
300: server
310: registration unit 320: database unit
330: Detection unit 331:
332: split image extracting unit 333: truth value determining unit
400: communication network

Claims (6)

delete A method for identifying a false print on a printed original printed with a halftone dot pattern together with an identification code, wherein the puzzled image is obtained by dividing an original image into a plurality of divided images,
(S100) for registering and storing, at a request of the registrant terminal, a contrast position for designating a divided image to be used for judging authenticity among the plurality of divided images of the puzzle image, which is a photographed image of the original print, ;
A step S200 of receiving a sensed image as a sensed image of a printed matter to be authenticated having a standard of the genuine printed matter, from a server according to a sensory request of a user terminal;
A genuine photographed image matching the identification code of the photographed image to be photographed is selected from the genuine photographed images registered in the server, and a divided image of the contrast position is extracted for each of the photographed image and the genuine photographed image, Determining (S300) whether or not the photographed object image is true or false, and notifying the determination result to the user terminal;
, ≪ / RTI &
At least the pattern represented in the divided image of the contrast position is designed to extend to an adjacent region in the original image,
The detecting step S300 further extracts the divided image adjacent to the divided image at the contrast position on the original image, and then applies the pattern continuity between the divided images to the determination of the authenticity
A method of detecting forged prints. 3. The method of claim 2,
In the determining step (S300), the pattern comparison includes a line thickness comparison
A method of detecting forged prints. 3. The method of claim 2,
In the original registration step S100, the plurality of contrast positions are designated, and the divided images of the contrast positions are designated to be adjacent to each other in the original image,
In the determining step S300, pattern comparison is performed for each divided image at a plurality of contrast positions and at the same time, restored to a position on the original image, adjacent to each other, and pattern continuity between the divided images is also compared to determine the authenticity of the object image
A method of detecting forged prints. The method according to any one of claims 2 to 4,
In the determining step S300, a comparison of halftone dots after the pattern comparison of the divided images at the contrast position is performed to determine whether the object image to be authenticated is authentic or not,
The halftone comparison is to compare at least one of the halftone dot spacing, the number of halftone dots, the halftone dot position,
A method of detecting forged prints. 6. The method of claim 5,
The original printed matter is halftone printed after a specific color is further applied to at least one of the boundary lines of the pagination image,
The specific color and the position of the divided boundary line on which the specific color is applied are registered in the server in the original registration step S100,
The identifying step S300 identifies and extracts a divided image of the contrast position using the color-based boundary line as a reference line based on the registered color and the position of the dividing boundary line, and compares the divided images
A method of detecting forged prints.

Images