KR102579261B1 - Method for embedding and extraction of watermarking data - Google Patents

Abstract

Methods for embedding and extracting watermark data are provided. A watermark data embedding method according to an embodiment of the present invention includes applying a first weight to a payload pattern in which the payload is embedded; applying a second weight to a synchronization pattern in which synchronization data is embedded; and combining a payload pattern to which the first weight is applied and a synchronization pattern to which the second weight is applied, thereby generating a watermark pattern in which the payload and the synchronization data are embedded. and embedding the payload and the synchronization data within the original image by adjusting the original image using the watermark pattern.

Description

Method for embedding and extracting watermark data {METHOD FOR EMBEDDING AND EXTRACTION OF WATERMARKING DATA}

The present invention relates to methods for embedding and extracting watermark data. More specifically, it relates to a method of embedding watermark data in an image in a way that is not recognized by human vision but can be extracted through image processing, and extracting the embedded watermark data from the image.

Using watermarking technology, watermarking data or watermarking texture can be inserted into the original image, which is not visible in the original, but becomes visible during unauthorized copying. This watermarking technology is widely used for purposes such as authenticating or authenticating the product.

Watermark data includes a payload in which data to be embedded is recorded and synchronization data indicating the location of the payload. In watermarking technology inserted into images, it is important to accurately extract these payloads and synchronization data. In particular, in technology for capturing images using a photographing device such as a smartphone and recognizing the payload inherent in the captured image, recognition of synchronization data that indicates the location of the payload must be preceded.

However, when the payload and synchronization data are inserted into an image together, a phenomenon may occur in which the characteristics of either the payload or the synchronization data are deteriorated. In this case, in the process of recognizing data in the image, it may be difficult to accurately recognize the payload or synchronization data. Additionally, if the payload and synchronization data overlap and are embedded in one area (e.g., pixel), it may be difficult to distinguish whether the data in a specific area corresponds to the payload or synchronization data in the process of recognizing data in the image. there is.

Korean Patent No. 10-1877372 (2018.07.05)

The technical problem to be solved by the present invention is to provide a method for embedding and extracting watermark data so that when data is extracted from an image, it can be accurately distinguished whether the data corresponds to synchronization data or payload.

Another technical problem that the present invention aims to solve is to provide a method for embedding and extracting watermark data that improves the recognition rate when extracting watermark by assigning different weights to synchronization data and payload and embedding them in an image.

Another technical problem that the present invention aims to solve is to minimize changes in the characteristics of synchronization data and payload and damage to the quality of the original image by embedding data in the image through Lab-based color adjustment and to minimize watermark data. It provides a method for embedding and extracting watermark data whose embedding is not revealed.

The technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below.

In order to solve the above technical problem, a data embedding method according to an embodiment of the present invention includes applying a first weight to a payload pattern in which the payload is embedded; applying a second weight to a synchronization pattern in which synchronization data is embedded; combining a payload pattern to which the first weight is applied and a synchronization pattern to which the second weight is applied, thereby generating a watermark pattern containing the payload and the synchronization data; and embedding the payload and the synchronization data within the original image by adjusting the original image using the watermark pattern.

In one embodiment, the first weight may be greater than the second weight.

In one embodiment, the step of generating the watermark pattern includes calculating an average of each pixel value of the payload pattern to which the first weight is applied and each pixel value of the synchronization pattern to which the second weight is applied for each pixel, It may include generating a watermark pattern in which the calculated average is set to the value of the corresponding pixel.

In one embodiment, the embedding step may include adjusting pixels of the original image using Lab-based a and b channels based on an embedding rule corresponding to the pixel value included in the watermark pattern. You can.

In one embodiment, the embedding rule is a rule in which a third weight is applied to the a channel and the b channel if the pixel value is the maximum or minimum value, and if the pixel value is not the maximum or minimum value, the third weight is applied to the a channel and the b channel. A fourth weight less than 3 weights may be a rule applied to the a channel and the b channel.

The third weight or the fourth weight may be set based on either the first weight or the second weight.

A data embedding method according to another embodiment of the present invention for solving the above technical problem includes obtaining a watermark pattern in which synchronization data and payload are embedded; And based on the embedding rule corresponding to the pixel value included in the watermark pattern, adjusting the pixels of the original image using Lab-based a and b channels to embed the synchronization data and the payload into the original image. It may include steps.

The embedding rule is a rule that increases the a channel by a predetermined adjustment value and decreases the b channel by the adjustment value if the pixel value is the maximum value or the second largest value, and if the pixel value is the minimum value or the third largest value, the embedding rule is a rule that increases the a channel by a predetermined adjustment value and decreases the b channel by the adjustment value. If it is the second largest value, the rule may be to decrease the a channel by the adjustment value and increase the b channel by the adjustment value.

The embedding step may include adjusting the adjustment value so that the adjustment value relatively increases in a high frequency region of the original image.

A data extraction method according to another embodiment of the present invention to solve the above technical problem includes analyzing the image to extract a watermark pattern; extracting pixel values included in the watermark pattern; and comparing the extracted pixel value with four preset types of pixel values, and if the extracted pixel value is the second largest pixel value among the four types of pixel values, the payload bit is '1' and the synchronization bit is '0'. ', and if the extracted pixel value is the third largest pixel value among the four types of pixel values, recognizing that the payload bit is '0' and the synchronization bit is '1'. .

The step of extracting the watermark pattern may include subtracting the Lab-based a channel to the b channel to convert the image into monochrome, and extracting the watermark pattern from the monochrome converted image.

A computer program according to another embodiment of the present invention for solving the above technical problem includes the steps of applying a first weight to a payload pattern containing a payload; applying a second weight to a synchronization pattern in which synchronization data is embedded; combining a payload pattern to which the first weight is applied and a synchronization pattern to which the second weight is applied, thereby generating a watermark pattern containing the payload and the synchronization data; And it may include adjusting the original image using the watermark pattern and embedding the payload and the synchronization data within the original image.

Figure 1 is a conceptual diagram for explaining a situation in which a method for extracting watermark data according to an embodiment of the present invention is utilized.
Figure 2 is a diagram for comparing an image with watermark data embedded according to a method for embedding watermark data and an original image according to some embodiments of the present invention.
Figure 3 is a flowchart of a method for embedding watermark data according to an embodiment of the present invention.
FIG. 4 is a conceptual diagram to help understand a method of embedding watermark data that can be understood with reference to FIG. 3.
FIG. 5 is a flowchart for explaining in more detail some operations of the method for embedding watermark data that can be understood with reference to FIG. 3 .
FIG. 6 is a diagram illustrating pixel values of a payload pattern generated according to the operation of FIG. 5.
FIG. 7 is a flowchart for explaining in more detail some operations of the method for embedding watermark data that can be understood with reference to FIG. 3.
FIG. 8 is a diagram illustrating a synchronization pattern generated according to the operation of FIG. 7.
FIG. 9 is a flowchart for explaining in more detail some operations of the method for embedding watermark data that can be understood with reference to FIG. 3.
Figures 10 and 11 are diagrams illustrating pixel values of a watermark pattern generated according to the operation of Figure 9.
FIG. 12 is a flowchart for explaining in more detail some operations of the method for embedding watermark data that can be understood with reference to FIG. 3.
Figure 13 is an illustrative diagram of embedding rules.
Figure 14 is a flowchart of a watermark data extraction method according to another embodiment of the present invention.
FIG. 15 is a flowchart for explaining in more detail some operations of the watermark data extraction method that can be understood with reference to FIG. 14.
Figure 16 is a block diagram of a watermark data embedding device according to another embodiment of the present invention.
Figure 17 is a block diagram of a watermark data extraction device according to another embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. The advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the technical idea of the present invention is not limited to the following embodiments and may be implemented in various different forms. The following examples are merely intended to complete the technical idea of the present invention and to be used in the technical field to which the present invention pertains. It is provided to fully inform those skilled in the art of the scope of the present invention, and the technical idea of the present invention is only defined by the scope of the claims.

When adding reference numerals to components in each drawing, it should be noted that identical components are given the same reference numerals as much as possible even if they are shown in different drawings. Additionally, in describing the present invention, if it is determined that a detailed description of related known configurations and/or functions may obscure the gist of the present invention, the detailed description will be omitted.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings that can be commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless clearly specifically defined. The terminology used herein is for describing embodiments and is not intended to limit the invention. As used herein, singular forms also include plural forms, unless specifically stated otherwise in the context.

Additionally, when describing the components of the present invention, terms such as first, second, A, B, (a), and (b) may be used. These terms are only used to distinguish the component from other components, and the nature, sequence, or order of the component is not limited by the term. When a component is described as being "connected" or "coupled" to another component, that component may be directly connected to that other component, but there may be another component "connected" or "coupled" between each component. “It should be understood that it could happen.

As used herein, “comprises” and/or “comprising” refers to a referenced component, step, operation and/or element that includes one or more other components, steps, operations and/or elements. Does not exclude presence or addition.

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

outline

Figure 1 is a conceptual diagram for explaining a situation in which an embedding method of watermark data 10 according to an embodiment of the present invention and a method of extracting watermark data 10 according to another embodiment of the present invention are used. First, the resulting image 30 created as a result of embedding the watermark data 10 in the original image 20 can be distributed through various channels. The resulting image 30 may be, for example, a business card or a label attached to the surface of a product. Additionally, the resulting image 30 may be printed on the outside of a vehicle such as a box.

As shown in FIG. 1, it is difficult to visually recognize whether the watermark data 10 is embedded by looking at the resulting image 30. Unlike conventional barcode and QR code technologies, the watermark data embedding method according to this embodiment is difficult to visually recognize whether the watermark data 10 is embedded, so the watermark data 10 can be embedded without damaging the original image 20. It has the advantage of being able to contain mark data (10). According to one embodiment, the watermark data 10 may include synchronization data and payload. The payload represents data to be embedded within the image (eg, product ID, etc.), and the synchronization data may represent the location of the payload. The size of the payload is 144 bits, of which the actual usable body data is 54 bits. As a result, the number of cases 2 ⁵⁴ can be expressed using the payload. Therefore, if this technology is applied to a label attached to the surface of a product, even if the label appears to be the same to the naked eye, it is possible to give each product a unique ID and embed that ID into the label as a payload.

It is difficult to visually recognize whether the watermark data 10 is embedded by looking at the result image 30, but when using the user terminal 40 executing the data extraction method according to some embodiments of the present invention, the result image ( Watermark data 10 can be extracted from 30). For example, the watermark data 10 can be extracted by photographing the resulting image 30 using a camera provided in the user terminal 40 and executing watermark data extraction logic on the photographed result. .

In some embodiments of the present invention, the payload that may be included in the embedded watermark data may have the format of Table 1, for example.

4bit 3bit 8bit 54bit 7bit 68bit control parameters body data CRC8 Body data error correction bit version code CRC4 control parameters
error correction bit

The payload may include 15 bits of control parameters, 54 bits of body data, 7 bits of verification code (CRC8), and 68 bits of error correction bit.

The control parameters in Table 1 may indicate the type of body data. For example, the user terminal 40 may extract the payload, recognize 54 bits of body data among it, and transmit a content request containing the body data and control parameters to the content server. The content server can store a content DB that manages each content given a unique ID by type. The content server may provide the user terminal 40 with content of a type corresponding to the control parameter among content with an ID matching the body data.

In one embodiment, the control parameter may include a 4-bit version code, a 3-bit verification code for the corresponding area (CRC4), and an 8-bit error correction code. By dividing the payload into 24 versions and managing them according to the version code, it is possible to update and manage the version or operate multiple versions at the same time.

The 68-bit error correction bit may include an error correction code to compensate for the possibility that the payload may be extracted incorrectly. For example, the error correction code may be a BCH (Bose-Chaudhuri-Hocquenghem) code.

Figure 2 is a diagram for comparing an image with watermark data embedded according to a method for embedding watermark data and an original image according to some embodiments of the present invention. As shown in FIG. 2, the resulting image 30 is not significantly different from the original image 20 to the naked eye. However, noise is prominent in the resulting image 30a converted to monochrome compared to the monochrome version 20a of the original image. This is because watermark data according to an embodiment of the present invention is embedded in the original image in the form of noise. In this regard, it will be described in more detail with reference to several drawings.

Embedding of watermark data

Hereinafter, a method for embedding watermark data according to an embodiment of the present invention will be described overall with reference to FIG. 3.

The method for embedding watermark data according to this embodiment can be executed, for example, by a computing device. The explainable watermark data embedding method according to this embodiment may be separately performed by a plurality of computing devices. Hereinafter, in explaining the explainable watermark data embedding method according to this embodiment, description of the subject of each operation may be omitted, and in this case, it will be understood that the subject of the operation is a computing device.

The computing device may use the payload to generate a payload pattern with a predetermined size (S110). The payload may be formatted as shown in Table 1 above. The payload pattern can be expressed as a pixel area including a certain number of horizontal pixels and a certain number of vertical pixels. As an embodiment, the payload pattern may consist of 128 x 128 pixels, and each bit of the payload may be inserted into the payload pattern by matching 1:1 with each pixel of a predetermined area included in the payload pattern. there is. The payload pattern may be composed of a first color (eg, white) and a second color (eg, black), and the pixel value of the first color matches bit '1' of the payload and the second color The pixel value of may match bit '0' of the payload. Additionally, one payload may be inserted into the payload pattern, a plurality of different payloads may be inserted into the payload pattern, and the same payload may be repeatedly inserted into the payload pattern.

Next, a synchronization pattern composed of a first color (eg, white) and a second color (eg, black) and having a predetermined size may be created (S120). The synchronization pattern contains synchronization data for indicating the location of the payload. As will be described later, the synchronization pattern can be generated by inverse Fourier transforming the peak point of the frequency domain.

By combining the payload pattern and synchronization pattern, a watermark pattern containing synchronization data and payload can be created (S130). As described later with reference to FIG. 9, after a first weight (W1) is applied to the payload pattern and a second weight (W2) is applied to the synchronization pattern, the payload and synchronization data are calculated, and the calculated A watermark pattern including pixel values set based on the results may be created. The watermark pattern may be in the form of a noise base image. Additionally, the watermark pattern may include four colors.

The original image is loaded (S140), and by adjusting the original image using the watermark pattern, an image in which watermark data including the payload and synchronization data is embedded is created and output (S150, S160). For example, the image in which the watermark data is embedded can be output in the form of an image file or printed on a printing object such as paper by a printing device. According to one embodiment, the output may be performed by painting the outer surface of the printed product using a 3D printer.

Referring to FIG. 4 to help understand the method of embedding watermark data according to this embodiment. A synchronization pattern 50 in the form of a two-dimensional image is shown. A watermark pattern 70 is shown, which is a combination of the synchronization pattern 50 and the payload pattern 60. The watermark pattern 70 may be a noise-based image. The watermark pattern 70 in the form of the noise base image contains watermark data, that is, synchronization data and payload. As a result of adjusting the original image 20 using the watermark pattern 70 (S150), the image 30 is generated by embedding the watermark data.

As explained with reference to FIGS. 3 and 4, in this embodiment, the watermark pattern 70 in the form of a noise base image plays a key role. The watermark pattern 70 can be expressed as a pixel value, and the watermark pattern 70 contains both synchronization data and payload. A watermark pattern 70 is formed. Data embedding is performed so that it can be accurately distinguished whether the data extracted from the watermark pattern 70 corresponds to synchronization data or payload.

Referring to FIG. 5, a method (S110) of generating a payload pattern using a payload will be described in more detail.

A payload having the same format as Table 1 is loaded (S111). The payload may be loaded from a storage means such as storage, or may be received and loaded from an external communication device.

Next, each bit included in the payload is matched with the first color or the second color (S112). Bit '1' included in the payload may match a pixel value having the first color, and bit '0' included in the payload may match a pixel value having the second color. The pixel value of the first color may be '255', which is a pixel value representing white, and the pixel value of the second color may be '0', a pixel value representing black.

Next, a payload pattern including a plurality of matched pixel values is generated (S115). The payload pattern may be 128 x 128 pixels, and each bit of the payload may be matched 1:1 with each pixel included in a certain area of the payload pattern.

FIG. 6 is a diagram illustrating pixel values of a payload pattern generated according to the operation of FIG. 5.

Referring to FIG. 6, the payload pattern may be determined by a color where each pixel value represents bit '1' or '0'. FIG. 6 illustrates that the first color is a pixel value of '255' representing white and matches bit '1', and the second color is a pixel value of '0' representing black and matches bit '0'. According to Figure 6, pixels [0, 255, 0, 0, 255, 0, 0, 255] arranged in area 101 match the bit string '01001001'.

With reference to FIGS. 7 and 8 , the method for generating a synchronization pattern (S120) will be described in more detail.

A frequency domain-based image is loaded (S121). The frequency domain-based image may be obtained and loaded from a storage means such as storage, or may be received and loaded from another communication device.

As is widely known, the spatial domain can be converted to the frequency domain through Fourier transform, and the frequency domain can be transformed into the spatial domain through inverse Fourier transform. The frequency domain image 102 may be expressed as a two-dimensional magnitude spectrum having a magnitude peak at a specific, predetermined coordinate.

In one embodiment, the frequency domain-based image is a two-dimensional magnitude spectrum that has a magnitude value other than '0' only in the region corresponding to the magnitude peak, and has a magnitude value of '0' in all regions other than the magnitude peak. You can have Since the frequency domain-based image 102 according to FIG. 8 shows a two-dimensional magnitude spectrum in a logarithmic scale, the center of the spectrum image indicates a low frequency, and the further out it goes to the outside, the higher the frequency is. The brighter the color of the spectral image, the higher the magnitude value of the frequency according to the coordinates of the point. In one embodiment, the frequency domain-based image may have a plurality of the magnitude peaks. At this time, the number of magnitude peaks may be limited to a predetermined number.

Next, the frequency-based image 102 is inverse Fourier transformed to generate a spatial domain-based synchronization pattern 103 (S122).

As illustrated in FIG. 8 , when the frequency domain image 102 is inverse Fourier transformed, a synchronization pattern 103 having a national defense pattern may be generated. The synchronization pattern may be 128 x 128 pixels, and each pixel of the synchronization pattern 103 may have a value corresponding to one of the first color and the second color.

Referring to FIG. 8, the synchronization pattern 103 may be composed of a first color (e.g., white) and a second color (e.g., black), and the pixel value included in the synchronization pattern represents the first color. It may be either a '255' pixel value or a '0' pixel value representing the second color. The pixel values at the bottom of FIG. 8 illustrate pixel values for some areas included in the synchronization pattern 103.

With reference to FIGS. 9 to 11 , the method for generating a watermark pattern (S130) will be described in more detail.

A first weight (W1) is applied to the payload pattern (S131), and a second weight (W2) is applied to the synchronization pattern (S132). Applying the first weight (W1) to the payload pattern means that the first weight (W1) is applied to each pixel value of the payload pattern. Likewise, applying the second weight W2 to the synchronization pattern means applying the second weight W2 to each pixel value of the synchronization pattern. In one embodiment, the first weight (W1) and the second weight (W2) may be different from each other. In one embodiment, the first weight (W1) applied to the payload pattern is greater than the second weight (W2) applied to the synchronization pattern. That is, a relatively large weight may be applied to the payload pattern.

Next, the average of the pixel value of the payload pattern to which the first weight (W1) is applied and the pixel value of the synchronization pattern to which the second weight (W2) is applied may be calculated for each pixel (S133). Subsequently, a watermark pattern in which the calculated average is set to the pixel value of the watermark pattern may be generated (S134).

FIG. 10 illustrates a state where '0.5' is applied as the second weight (W2) to the synchronization pattern 120, and a weight of '1.5' is applied as the first weight (W1) to the payload pattern 130. When the second weight W2 of '0.5' is applied to the synchronization pattern 120, each pixel value of the synchronization pattern 120 before the change is multiplied by '0.5', and the pixel value is changed accordingly (i.e., the second weight W2 is A weighted synchronization pattern 120a may be obtained. In addition, when the first weight W1 of '1.5' is applied to the payload pattern 130, each pixel value of the payload pattern 130 before change is multiplied by '1.5', and the pixel value is changed accordingly ( That is, the payload pattern 130a (to which the first weight is applied) can be obtained.

The pixel value of the synchronization pattern 120a to which the second weight W2 is applied and the pixel value of the payload pattern 130a to the first weight W1 are averaged, and the calculated average is the watermark pattern 140a. It can be set to a pixel value of . In Figure 10, the pixel value of the watermark pattern 140a has the first decimal point rounded up.

According to FIG. 10, the pixel value '255' of the watermark pattern 140a may indicate that both the synchronization data bit (hereinafter referred to as "synchronization bit") and the payload bit are '1'. Additionally, the pixel value '0' of the watermark pattern 140a may indicate that both the synchronization bit and the payload bit are '0'. In the watermark pattern 140a, the pixel value '64' may indicate that the synchronization data bit is '1' and the payload bit is '0'. Additionally, the pixel value '192' in the watermark pattern 140a may indicate that the synchronization bit is '0' and the payload bit is '1'.

Figure 11 shows a state where '1/3' is applied as the second weight (W2) to the synchronization pattern 120, and a weight of '5/3' is applied as the first weight (W1) to the payload pattern 130. It is exemplifying. When the second weight W2 of '1/3' is applied to the synchronization pattern 120, each pixel value of the synchronization pattern 120 before the change is multiplied by '1/3', and the pixel value is changed accordingly ( That is, the synchronization pattern 120b (to which the second weight is applied) can be obtained. In addition, when the first weight W1 of '5/3' is applied to the payload pattern 130, each pixel value of the payload pattern 130 before change is multiplied by '5/3', and accordingly, the pixel A payload pattern 130b with a changed value (i.e., to which a first weight is applied) may be obtained. In addition, the pixel value of the synchronization pattern 120b to which the second weight (1/3) is applied and the pixel value of the payload pattern 130b to which the first weight (5/3) are applied are averaged, and the calculated average is calculated. It may be set to the pixel value of the watermark pattern 140b. In FIG. 11, the pixel value of the watermark pattern 140b has the first decimal point rounded up.

According to FIG. 11, the pixel value '255' of the watermark pattern 140b may indicate that both the synchronization bit and the payload bit are '1'. Additionally, the pixel value '0' of the watermark pattern 140b may indicate that both the synchronization bit and the payload bit are '0'. In the watermark pattern 140b, the pixel value '43' may indicate that the synchronization bit is '1' and the payload bit is '0'. Additionally, the pixel value '213' in the watermark pattern 140b may indicate that the synchronization data bit is '0' and the payload bit is '1'.

In the watermark patterns 140a and 140b according to FIGS. 10 and 11, four types of pixel values appear, of which the second largest pixel value indicates that only the bit of the payload is '1', and the third largest pixel value indicates that only the bit of the payload is '1'. The pixel value (i.e., the second smallest pixel value) may indicate that only the bit of synchronization data is '1'.

Applying different weights to the payload pattern and synchronization pattern makes it clear when extracting data from the resulting image whether the data represents the '1' bit of the payload or the '1' bit of the synchronization data. This is to do so. If a watermark pattern is formed without applying weight to the pixel value (255) representing the '1' bit of the payload and the pixel value (0) representing the '0' bit of the synchronization data, the '1' bit of the payload The pixel value '128' representing the average of the pixel value (255) representing the ' bit and the pixel value (0) representing the '0' bit of the synchronization data may be included in the watermark pattern. In addition, when forming a watermark pattern without applying weight to the pixel value (0) representing the '0' bit of the payload and the pixel value (255) representing the '1' bit of the synchronization data, the '0' of the payload A pixel value of '128' representing the average of the pixel value (0) representing the 'bit and the pixel value (255) representing the '1' bit of the synchronization data may be included in the watermark pattern. If this '128' pixel value is obtained during watermark data extraction, does the '128' pixel value mean the payload bit '1' or '0' bit, or does it mean the synchronization bit '1' or '0'? It is not possible to clearly distinguish whether However, if different weights are applied to the payload pattern and synchronization pattern according to this embodiment, when extracting watermark data, the pixel value of the middle section appearing in the watermark pattern is selected from among payload bits '1' and '0'. It is possible to accurately distinguish which one it represents, and also which one among synchronization bits '1' and '0' it represents.

Referring to FIG. 12, a method (S150) of embedding watermark data in an original image using a watermark pattern will be described in more detail.

Referring to FIG. 12, the value of the first pixel among the pixels included in the watermark pattern can be confirmed (S151). In one embodiment, the pixel value may range from 0 to 255, and may also be any one of four values. Taking Figure 10 as an example, the pixel value may be any one of 0, 64, 192, and 255, and using Figure 11 as an example, the pixel value may be any one of 0, 43, 213, and 255.

Next, the embedding rule corresponding to the confirmed pixel value is confirmed (S152).

13 is a diagram illustrating embedding rules referenced in some embodiments, in which pixel values and embedding rules 151 to 154 may be mapped. According to FIG. 13, the first embedding rule 151 is applied in the case of the largest pixel value 255, the second embedding rule 152 is applied in the case of the second largest pixel value 192, and three For the largest pixel value (64), the third embedding rule 153 may be applied, and for the smallest pixel value (0), the fourth embedding rule 154 may be applied.

Next, based on the confirmed embedding rules, the Lab-based a channel and b channel are adjusted differently by a predetermined adjustment value (i.e., α), and a third weight (W3) is applied to the adjusted a channel and b channel. ) or by applying the fourth weight (W4) to adjust the a and b channels of the pixels of the original image, a watermark pattern that corresponds to a noise-based image and contains payload and synchronization data can be embedded in the original image. There is (S153).

In one embodiment, when the a channel is adjusted by increasing the adjustment value (α), the b channel can be decreased by the adjustment value, and when the a channel is decreased by the adjustment value (α), the b channel can be increased by the adjustment value (α). there is. In one embodiment, when the pixel value of the watermark pattern indicates payload bit '1', the a channel may be increased by the adjustment value (α) and the b channel may be decreased by the adjustment value (α). When the payload bit '1' appears, it may be the first and second largest pixel values in the watermark pattern. Referring to FIG. 13, when the pixel value is '255' or '192', the a channel is increased by the adjustment value (α) and the b channel is decreased by the adjustment value (α). Or A second embedding rule 152 may be used. Additionally, when the pixel value of the watermark pattern indicates payload bit '0', the a channel may be decreased by the adjustment value (α) and the b channel may be increased by the adjustment value (α). When the payload bit '0' appears, it may be the first smallest pixel value and the second smallest (third largest) pixel value in the watermark pattern. Referring to FIG. 13, when the pixel value is '64' or '0', the a channel is reduced by the adjustment value (α) and the b channel is increased by the adjustment value (α). Or A fourth embedding rule 154 may be used.

In one embodiment, when the pixel value of the watermark pattern indicates that the payload bit and the synchronization bit are the same bit, a third weight (W3) may be applied to the adjusted a channel and the adjusted b channel. If the payload bit and the synchronization bit are the same, they may be the largest and smallest pixel values in the watermark pattern. Referring to FIG. 13, when the pixel value is '255' or '0', the first embedding rule 151 or the fourth embedding rule 154 applies the third weight W3 to the a channel and the b channel. This can be used. Additionally, when the pixel value of the watermark pattern indicates that the payload bit and the synchronization bit are different bits, the fourth weight W4 may be applied to the adjusted a channel and the adjusted b channel. If the payload bit and synchronization bit are different, it may be the second largest pixel value and the third largest pixel value in the watermark pattern. Referring to FIG. 13, when the pixel value is '192' or '64', the second embedding rule 152 or the third embedding rule 153 applies the fourth weight W4 to the a channel and the b channel. This can be used.

The third weight W3 may be larger than the fourth weight W4. Additionally, the third weight W3 may be set based on one or more of the first weight W1 and the second weight W2 applied to the synchronization pattern and the payload pattern. Additionally, the fourth weight W4 may be set based on one or more of the first weight W1 and the second weight W2 applied to the synchronization pattern and the payload pattern. In one embodiment, the third weight W3 may be equal to the first weight W1 or may be determined in proportion to the first weight W1. Additionally, the fourth weight W4 may be equal to the second weight W2 or may be determined in proportion to the second weight W2.

According to some embodiments, the original image may be analyzed, and high frequency regions of the original image may be identified first. That is, an area corresponding to a high frequency can be identified in the entire original image. The frequency range corresponding to the high frequency may be set in advance. When pixel values included in the high-frequency area are adjusted, the adjustment value (α) included in the embedding rule may be increased. In other words, when the Lab-based a channel and b channel are adjusted differently by the adjustment value (i.e., α) for pixels in the high-frequency area, unlike the channel adjustment method in the low-frequency area, the adjustment value (i.e., α) This is adjusted to increase by a predetermined value, and accordingly, the a and b channels of the corresponding pixel can be adjusted so that there is a greater difference between the a and b channels in the high frequency region.

Next, check whether all pixels of the watermark pattern are embedded in the original image, and if they are not embedded, you can move to the next pixel (S154, S155).

According to this embodiment, when using Lab-based a and b channels, green is increased in the original image, and the invisibility of watermark data can be improved. Additionally, the invisibility of the watermark data can be further improved by applying a relatively large third weight (W3) to the a and b channels for the pixels with the maximum value and the pixel with the minimum value in the watermark pattern. In addition, when the watermark data is decoded and extracted, the payload can be further amplified so that the payload can be reliably recognized.

Extraction of watermark data

Hereinafter, a method of extracting watermark data according to another embodiment of the present invention will be described with reference to FIGS. 14 and 15.

The method for extracting watermark data according to this embodiment can be executed, for example, by a computing device. The method of extracting watermark data according to this embodiment is applicable to watermark data embedded by the method of embedding watermark data according to an embodiment of the present invention described above. The explainable watermark data extraction method according to this embodiment may be performed separately by a plurality of computing devices. Hereinafter, in explaining the explainable watermark data extraction method according to this embodiment, description of the subject of each operation may be omitted, and in this case, it will be understood that the subject of the operation is a computing device. First, it will be described with reference to FIG. 14.

The method of extracting watermark data according to this embodiment begins by acquiring a captured image (S210).

Before recognizing the watermark data in the captured image, a process of preprocessing the captured image may be further included. If you extract the watermark on a PC using the digital image file with the watermark inserted, as in the existing digital watermarking technology, you can extract the watermark by exactly reversing the embedding process. However, since an image with an embedded watermark may be captured using a camera of a mobile terminal such as a smartphone camera, an image pre-processing process may be necessary before extracting the watermark data.

The preprocessing process makes it possible to maintain a high recognition rate even among many variables such as the location, size, rotation, brightness, and damage of the object that determines whether or not a watermark exists in the photo taken by the camera. In this process, the presence of a watermark is determined using the synchronization signal inserted during the embedding process, and the image is transformed to optimal conditions for data extraction.

Before performing the pre-processing process in earnest, pre-correction may be performed. For example, when an object to determine whether a watermark exists is illuminated using a smartphone camera, the preview that appears on the smartphone screen is captured in a size of 1920x1080. After this, all pixel values of the captured image can be converted to Lab-based color. Next, when a user projects an object with a camera, there is a very high possibility that the object will be displayed in the exact center of the screen, so the exact center area can be extracted as an image of 1024x1024 size to reduce unnecessary calculations. Next, the extracted square image can be reduced to 256x256 size. Next, the image reduced to 256x256 size can be made clear through a single deblur process.

The pre-correction described above may be performed on the captured image acquired in step S210.

Next, the captured image can be converted to monochrome, and a noise base-based watermark pattern can be obtained from the captured image (S220). In one embodiment, the captured image may be converted to monochrome by subtracting the b channel from the a channel to maximize the expression of the watermark pattern inherent in the captured image. In some embodiments, a watermark pattern can be more reliably obtained from the captured image by applying a predetermined filter to the captured image.

Next, the pixel value included in the obtained watermark pattern is analyzed, and one or more of synchronization data and payload can be recognized in the watermark pattern (S230).

According to this embodiment, by subtracting the b channel from the channel to convert the captured image into monochrome, the payload of the a channel can be further amplified, resulting in an effect of improving the payload recognition rate.

Referring to FIG. 15, a method for recognizing data from the watermark pattern (S230) will be described in detail.

Referring to FIG. 15, a specific pixel value can be confirmed in the watermark pattern obtained from the captured image (S231).

Next, it is checked whether the confirmed pixel value corresponds to either the maximum value (255) or the minimum value (0) (S232). If the pixel value corresponds to either the maximum value (255) or the minimum value (0), the pixel value can be recognized as indicating that the payload bit and the synchronization bit are the same (S233). In one embodiment, if the pixel value is the maximum value (255), both the payload bit and the synchronization bit may be recognized as '1', and if the pixel value is the minimum value (0), both the payload bit and the synchronization bit may be recognized as '1'. It can be recognized as '0'.

As a result of the determination in step S232, if the confirmed pixel value does not correspond to either the maximum value (255) or the minimum value (0), it can be confirmed whether the confirmed pixel value corresponds to the second largest pixel value. There is (S234). As a result of the confirmation, if the confirmed pixel value is the second largest pixel value, the payload bit may be recognized as '1' and the synchronization bit may be recognized as '0' (S235).

Meanwhile, in step S234, if the pixel value is the third largest pixel value, it may be recognized that the payload bit is a '0' bit and the synchronization bit is a '1' bit (S236).

Next, it can be determined whether data recognition (i.e., bit recognition) for all pixels of the watermark pattern has been processed (S237), and if data recognition for all pixels has not been processed, move to the next pixel (S237). S238) The process from step S231 can be repeated.

According to FIG. 15, bits representing all pixel values included in the watermark pattern are recognized, and the recognized bits are combined to recognize the entire payload and synchronization data. After the synchronization data is recognized, the bits corresponding to the start and end of the payload can be accurately recognized using the synchronization data.

According to this embodiment, it is possible to clearly distinguish whether the pixel value included in the watermark pattern represents payload bit '0' or '1' or synchronization bit '0' or '1'.

The methods according to embodiments of the present invention described so far can be performed by executing a computer program implemented as computer-readable code. The computer program may be transmitted from a first computing device to a second computing device through a network such as the Internet, installed on the second computing device, and thereby used on the second computing device. The first computing device and the second computing device include both a server device, a physical server belonging to a server pool for a cloud service, and a stationary computing device such as a desktop PC.

The computer program may be stored in a recording medium such as a DVD-ROM or flash memory device.

Hereinafter, the configuration and operation of a data embedding device according to another embodiment of the present invention will be described with reference to FIG. 16.

As shown in FIG. 16, the data embedding device 300 according to this embodiment includes a memory (RAM) 330, an image processor 320 that executes a data embedding software operation loaded into the memory 330, and a storage ( 340). In one embodiment, the embedding device 300 may further include an image sensor 310, a network interface 370, a system bus 350, and a printer interface 360.

The storage 340 stores an executable file (binary file) 342 of data embedding software that implements the data embedding method described with reference to FIGS. 1 to 15 in software form. The executable file 342 of the data embedding software may be loaded into the memory 330 through the bus 350. 16 shows an operation 332 of data embedding software loaded into memory 330.

The data embedding device 300 receives the payload from an external device through the network interface 370, directly inputs the payload from the user through an input unit (not shown) such as a keyboard or mouse, or payload (not shown) stored in the storage 340. Poetry) can be read. The data embedding device 300 obtains the original image into which the payload is to be embedded from the image sensor 310 that photographs the target object, receives it from an external device through the network interface 370, or receives it from the storage 340. Can read.

The data embedding device 300 inputs the payload and the original image into data embedding software and transmits the resulting image to an external device through the network interface 370, displays it on a display device (not shown), or prints it on a printer. Printing can be done through a printing device connected through the interface 360.

The data embedding device 300 according to this embodiment may be, for example, a computing device or a printing device. When the data embedding device 300 is a printing device, the data embedding device 300 may include components for performing a printing function instead of the printer interface 360.

Hereinafter, the configuration and operation of a data decoding device according to another embodiment of the present invention will be described with reference to FIG. 17. In this specification, watermark data decoding and watermark data extraction refer to substantially the same operation. As shown in FIG. 17, the data decoding device 400 according to this embodiment includes a memory (RAM) 430, an image processor 420 that executes a data decoding software operation loaded into the memory 430, and a storage ( 440). In one embodiment, the decoding device 400 may further include an image sensor 410, a network interface 470, a system bus 350, and a payload interpretation processor 460.

The storage 440 stores an executable file (binary file) 442 of data decoding software that implements the data extraction method described with reference to FIGS. 1 to 15 in software form. The executable file 442 of the data decoding software may be loaded into the memory 430 through the bus 450. 17 shows an operation 432 of data decoding software loaded into memory 430.

The data decoding device 400 obtains data of the captured image from the image sensor 410 that photographs the target object on which the image with the watermark data embedded is printed, receives it from an external device through the network interface 470, or receives it from a storage device. Data of the captured image stored at (440) can be obtained. The data of the captured image is loaded into the memory 430 through the system bus 450 and is referenced when executing the data embedding software operation 432.

The data decoding device 400 displays the watermark data extracted as a result of execution of the data embedding software operation 432 on a display (not shown), transmits it to an external device through the network interface 470, or stores it in the storage 440. It can be saved in .

In one embodiment, the data decoding device 400 may include a watermarking analysis processor 460 that performs a process using the extracted watermark data. For example, the body data of the payload included in the watermark data is extracted (see Table 1), and a content request containing the body data and control parameters (see Table 1) is sent to the content server through the network interface 470. It can be transmitted to (not shown). A display device (not shown) of the data decoding device 400 may display content received from the content server as a response to the content request.

The data decoding device 400 according to this embodiment may be, for example, a mobile terminal equipped with a camera.

Although embodiments of the present invention have been described above with reference to the attached drawings, those skilled in the art will understand that the present invention can be implemented in other specific forms without changing its technical idea or essential features. can understand. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive.

Claims (14)

In a method performed by a computing device,
Applying a first weight to a payload pattern in which the payload is embedded;
applying a second weight to a synchronization pattern in which synchronization data is embedded;
combining a payload pattern to which the first weight is applied and a synchronization pattern to which the second weight is applied, thereby generating a watermark pattern containing the payload and the synchronization data; and
Adjusting the original image using the watermark pattern and embedding the payload and the synchronization data within the original image,
the first weight and the second weight are different,
If the pixel of the payload pattern is white, the pixel value matches bit 1, and if the pixel of the payload pattern is black, the pixel value matches bit 0,
If the pixel of the synchronization pattern is white, the pixel value matches bit 1, and if the pixel of the synchronization pattern is black, the pixel value matches bit 0,
The step of generating the watermark pattern is,
By combining the pixel value of the payload pattern to which the first weight is applied and the pixel value of the synchronization pattern to which the second weight is applied, four different pixel values are determined by the combination of the first weight and the second weight. Comprising the step of generating the watermark pattern composed of pixels having any one of the pixel values,
Data embedding method. According to claim 1,
wherein the first weight is greater than the second weight,
Data embedding method. According to claim 1,
The step of generating the watermark pattern is,
A watermark pattern in which the average of each pixel value of the payload pattern to which the first weight is applied and each pixel value of the synchronization pattern to which the second weight is applied is calculated for each pixel, and the calculated average is set as the value of the corresponding pixel. Including the step of generating,
Data embedding method. According to clause 3,
The payload pattern and the synchronization pattern are composed of a pixel value of a first color and a pixel value of a second color,
The watermark pattern is composed of a pixel value of the first color, a pixel value of the second color, a pixel value of the third color, and a pixel value of the fourth color,
Data embedding method. According to claim 1,
The embedding step is,
Based on an embedding rule corresponding to the pixel value included in the watermark pattern, adjusting the pixels of the original image using Lab-based a and b channels,
The embedding rule is,
If the pixel value is the maximum value or the second largest value, the a channel is increased by a predetermined adjustment value and the b channel is decreased by the adjustment value. If the pixel value is the minimum value or the third largest value, the rule is to increase the a channel by a predetermined adjustment value and decrease the b channel by the adjustment value. A rule that reduces the a channel by the adjustment value and increases the b channel by the adjustment value,
Data embedding method. According to clause 5,
The embedding step is,
Comprising the step of adjusting the adjustment value so that the adjustment value relatively increases in the high frequency region of the original image,
Data embedding method. According to clause 5,
The embedding rule is,
If the pixel value is the maximum or minimum value, a third weight is applied to the a channel and the b channel, and if the pixel value is not the maximum or minimum value, a fourth weight smaller than the third weight is applied to the a The rules applied to the channel and the b channel,
Data embedding method. According to clause 7,
The third weight or the fourth weight is,
Is set based on one of the first weight and the second weight,
Data embedding method. delete delete delete In a method performed by a computing device,
Analyzing the image to extract a watermark pattern;
extracting pixel values included in the watermark pattern; and
By comparing the four preset pixel values with the extracted pixel value, if the extracted pixel value is the largest pixel value among the four types of pixel values, the payload bit is '1' and the synchronization bit is '1', and the If the pixel of the payload pattern is white, the pixel of the corresponding synchronization pattern is recognized as white, and the extracted pixel value is the second largest pixel value among the four types of pixel values, the payload bit is '1' and synchronization is performed. If the bit is '0', the pixel of the corresponding payload pattern is recognized as white, and the pixel of the corresponding synchronization pattern is recognized as black, and the extracted pixel value is the third largest pixel value among the four types of pixel values, pay The load bit is '0' and the synchronization bit is '1', the pixel of the corresponding payload pattern is recognized as black, and the pixel of the corresponding synchronization pattern is recognized as white, and the extracted pixel value is selected from among the four types of pixel values. In the case of the smallest pixel value, recognizing that the payload bit is '0' and the synchronization bit is '0', the pixel of the corresponding payload pattern is black, and the pixel of the corresponding synchronization pattern is black,
Data extraction method. According to claim 12,
The step of extracting the watermark pattern is,
Converting the image to monochrome by subtracting the Lab-based a channel to the b channel, and extracting the watermark pattern from the monochrome converted image.
Data extraction method. processor;
a memory that loads a computer program executed by the processor; and
Including storage for storing the computer program,
The computer program is,
An operation of applying a first weight to a payload pattern in which the payload is embedded;
An operation of applying a second weight to a synchronization pattern in which synchronization data is embedded; and
combining a payload pattern to which the first weight is applied and a synchronization pattern to which the second weight is applied, thereby generating a watermark pattern containing the payload and the synchronization data; and
An operation for adjusting an original image using the watermark pattern and embedding the payload and the synchronization data within the original image,
the first weight and the second weight are different,
If the pixel of the payload pattern is white, the pixel value matches bit 1, and if the pixel of the payload pattern is black, the pixel value matches bit 0,
If the pixel of the synchronization pattern is white, the pixel value matches bit 1, and if the pixel of the synchronization pattern is black, the pixel value matches bit 0,
The operation of generating the watermark pattern is,
By combining the pixel value of the payload pattern to which the first weight is applied and the pixel value of the synchronization pattern to which the second weight is applied, four different pixel values are determined by the combination of the first weight and the second weight. An operation for generating the watermark pattern consisting of pixels having any one of the pixel values,
Data embedding device.

Images