KR20090104349A - Wartermark insertion/detection apparatus and method thereof - Google Patents

Abstract

PURPOSE: A wartermark insertion/detection apparatus and a method thereof are provided to increase the information protection efficiency remarkably by inserting multiple bits through the division of a region based on a pixel strength value. CONSTITUTION: A wartermark insertion/detection apparatus(100) includes a frame arrangement unit(110), a frame region split unit(120) and a bit information insertion unit. The frame arrangement unit arranges a reception image frame in input order, and generates a first group to include the arranged image frame as many as the set number. The frame region split unit splits each frame into at least two regions based on the strength value of the pixels included in the each image frame. The bit information insertion unit inserts bit information, which is matched with the user information, into the designated region among the split regions.

Description

Watermark inserting / extracting device and method thereof {WARTERMARK INSERTION / DETECTION APPARATUS AND METHOD THEREOF}

The present invention relates to a watermark embedding / extraction apparatus and a method thereof. In particular, the present invention relates to an apparatus and method for inserting / extracting a watermark in video content to protect the copyright of the creator of the video content.

With the development of technology, various types of video information are changed from analog to digital information and are widely used through the Internet or storage media.

Digital image information has advantages such as convenience of storage and conversion, usefulness of mass copying and distribution, whereas lack of security device to protect intellectual property rights of image information, intellectual property rights over internet and virtual space (storage media, etc.) Infringement is getting serious day by day.

Various security devices are currently being developed to solve such infringement of intellectual property rights.

As a representative method for solving the intellectual property infringement of the video information is the encryption method (DRM). This is a method of encrypting the transmitted data and transmitting the encrypted data so that the receiver can enter the secret key to restore the encrypted data, so that even if a third party illegally receives the data during transmission, it cannot be restored to the original data. However, the encryption method (DRM) has a problem that the copyright protection of the data itself is impossible because legitimate recipients can disclose data similar to the original extracted by a method such as video capture to a third party.

Accordingly, in order to solve such a problem, a watermarking method has emerged, in which digital data is invisibly structured and inserted into original data to enable copyright protection of the data itself.

Watermarking methods have recently been applied to various applications such as graphics, communication, and audio, and their importance is increasing day by day.

The watermarking method is a technique for protecting intellectual property by inserting copyright holder's information (user information) into the original data for materials having intellectual property protection characteristics that can be distributed on the network. It is divided into blind watermarking requiring data and non-blind watermarking requiring no original data.

The watermark used in watermarking must satisfy the following conditions.

1. Robustness. The watermark information inserted in the original image should be extracted without being damaged by an external attack.

2. There must be invisibility. For the content in which the watermark is inserted in the original data, the user should not feel the difference from the original data. That is, by inserting a watermark, the original data should not be damaged to the extent that the user recognizes the insertion of the watermark.

Conventional watermarking techniques used for transmitting and receiving image information have been based on image watermarking techniques. Therefore, the image is divided into single images, and an image watermark is inserted into each frame.

Representative image watermark embedding methods include a spatial method and a frequency domain method.

The spatial method is a method of inserting a watermark by making a small change in the pixel value of the original image in terms of space. The problem with this method is that in the process of applying the image watermark, various geometric deformation attacks can be made in the case of the image information, and thus, geometrical sync needs to be met. Currently, various studies are in progress, but there are many difficulties in the actual implementation. In addition, since a specific pattern encrypted is inserted, this is very different from the original image, and thus there is a problem that is noticeable to the user.

The frequency domain method converts digital data into analog signals of frequency components and converts and inserts watermarks using the same conversion method. The problem with this method is that the watermark must be inserted in the specific frequency region of the image information. If the watermark is inserted at an excessively low frequency, the watermark is recognized by the user's eyes, and thus the invisibility is lowered. In this case, since the watermark disappears in the compression process such as converting the codec, the robustness is reduced.

On the other hand, there is a conventional spatial method and frequency domain method in which a multi-bit is inserted into one image or frame by applying the spatial method and the frequency domain method.

The spatial method of inserting the multi-bits will be described with reference to FIG. 1 below.

1 is a diagram illustrating a segmentation of an image image by using a spatial method of inserting a multi-bit according to the prior art.

The spatial method of inserting a multi-bit is a method of dividing a plurality of areas into a frame based on a geometric reference as shown in FIG. 1, and it is possible to simply insert a plurality of bits of information into a frame, but there are some problems.

First, to divide the area by geometric reference, it is necessary to match the geometric sync. This makes it vulnerable to geometric attacks such as shooting with a camcorder, image rotation, cropping, and translation, and requires another additional algorithm to solve the problem. In addition, when different bit information is inserted into adjacent areas, there is a problem in that the invisibility of the watermark is inferior because the difference in value is increased around the boundary line.

The frequency domain method of inserting multiple bits has a problem in that information of a corresponding region disappears completely when a specific frequency region is attacked due to a method of inserting different bits of information into various frequency regions, compression, and filtering.

Therefore, there is a need for a new watermarking technology that can enhance the efficiency of copyright protection by inserting / extracting watermarks with increased robustness and invisibility.

It is an object of the present invention to provide an apparatus and method for inserting / extracting a watermark in video content to protect the copyright of the creator of the video content.

According to a feature of the present invention for solving the above problems, the watermark embedding apparatus for embedding a watermark in the image information,

A frame arrangement unit for arranging the received image frames in an input order and generating a first group such that the arranged image frames are included in a set number; A frame region dividing unit which divides each image frame into two or more regions based on intensity values of pixels included in each image frame in the first group; And a bit information inserting unit inserting bit information matching the user information into a designated area among the divided areas.

The bit information inserting unit inserts the bit information by changing a color component of the designated area in a previous or subsequent image frame based on the reference image frame in the first group.

Here, the bit information inserter inserts the bit information by increasing or decreasing the color components by a predetermined number.

According to a feature of the invention, the watermark extraction apparatus for receiving the watermark embedded image information and extracting the watermark,

A frame arrangement unit for arranging image frames in an input order and generating a first group such that the arranged image frames are included in a set number; A frame region dividing unit which divides each image frame into two or more regions based on intensity values of pixels included in each image frame in the first group; And a bit information extraction unit for extracting bit information from a designated area among the divided areas.

Here, the bit information extraction unit,

Bit information is extracted by comparing color components of a specified region of a previous image frame and a subsequent image frame based on the reference image frame in the first group.

The apparatus for extracting watermarks may further include a converting unit for converting RGB (Red, Green, Blue) components included in each image frame in the first group into U region values on YUV.

The bit information extracting unit extracts the bit information by comparing a U region average value of the designated region of the previous image frame and the subsequent image frames based on the reference image frame in the first group.

According to a feature of the invention, the watermark embedding method of the apparatus for embedding the watermark in the image information,

Arranging image frames in an input order to generate a first group such that the image frames are included in a set number; Dividing each image frame into two or more regions based on pixel intensity values of each image frame in the first group; And inserting bit information about the user information into a designated area among the divided areas of each of the image frames.

Here, the watermark embedding method further comprises the step of receiving the user information from the authentication unit for performing authentication for the user.

According to a feature of the invention, the watermark extraction method of the apparatus for extracting the watermark by receiving the image information embedded with the watermark,

Arranging image frames in an input order to generate a first group such that the image frames are included in a set number; Dividing each image frame into two or more regions based on pixel intensity values of each image frame in the first group; And extracting bit information from a designated area among the divided areas.

Here, the step of extracting the bit information,

The bit information is extracted by comparing color components of a specified region of a previous image frame and a subsequent image frame based on the reference image frame in the first group.

According to the present invention, it is possible to minimize the change of the original image of the video content can be expected to effectively prevent the acts such as illegal copying, distribution of the video content.

In addition, by inserting multiple bits by dividing the region based on the pixel intensity value, the effect of significantly increasing the information protection rate can be expected, and information loss due to compression and filtering can be minimized.

In addition, since the bit is inserted based on the pixel intensity value, unlike the conventional multi-bit insertion, it is possible to expect an effect of significantly increasing invisibility and robustness by eliminating the difference in the peripheral value of the boundary line.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and like reference numerals designate like parts throughout the specification.

Throughout the specification, when a part is said to "include" a certain component, it means that it can further include other components, without excluding other components unless specifically stated otherwise. In addition, the terms “… unit”, “… unit”, “module”, etc. described in the specification mean a unit that processes at least one function or operation, which may be implemented by hardware or software or a combination of hardware and software. have.

Now, a watermark embedding / extraction apparatus and a method thereof according to an embodiment of the present invention will be described with reference to the drawings.

2 is a block diagram illustrating a watermarking system according to an exemplary embodiment of the present invention.

As shown in FIG. 2, a watermarking system according to an exemplary embodiment of the present invention includes a watermark embedding apparatus 100 and a watermark extracting apparatus 200.

The watermark embedding apparatus 100 includes a frame arranging unit 110, a frame region dividing unit 120, a YUV converting unit 130, and a bit information inserting unit 140.

The frame arranging unit 110 receives image information from an apparatus for providing image information, such as an upper image processing unit (omitted) or a storage unit storing image information (omitted), and inputs an image frame of the received image information. Arrange them in the order in which they occur. Here, it is assumed that the image information is RGB (Red, Green, Blue; hereinafter referred to as "RGB") (hereinafter also referred to as "color component").

In addition, the frame arranging unit 110 divides the arranged image frames into groups of 2n + 1 (where n = 1, 2, and 3 are natural numbers).

The frame region dividing unit 120 sequentially receives image frames in group units from the frame arranging unit 110 and divides the frame region for each image frame in the received group. In this case, the frame region dividing unit 120 divides the region by using a histogram distribution of intensity values I of pixels using RGB of the image frame. For example, when one image frame is divided into four regions, the first region is 171 or more, the second region is 141 to 170, the third region is 101 to 140, and the fourth region is based on the intensity values of the pixels. The area is less than 100.

Herein, the pixel intensity value is represented by Equation 1 below.

픽셀 강도값(I) = 혹은 (R+G+B)/3

Pixel intensity value (I) = or (R + G + B) / 3

By inserting the bit information by dividing the image frame area based on the intensity of the pixel, there is a great advantage that can transmit a lot of bit information to the watermark extraction apparatus 200 in a state of minimizing the original image change.

Such a frame region divided by the frame region divider 120 will be described later with reference to FIGS. 3 and 4.

The YUV converter 130 converts the RGB of the image frame received from the frame region divider 120 into YUV and provides the converted YUV to the bit information inserter 140. Here, among the YUVs converted by the YUV converter 130, Y represents brightness and UV represents color.

At this time, the YUV converter 130 converts RGB to YUV using Equation 2 below.

Y = 0.299R + 0.587G + 0.114B

U = -0.169R-0.332G + 0.5B

V = 0.5R-0.419G-0.0813B

The bit information inserting unit 140 receives user information of an authenticated user from a higher level such as an authentication unit (not shown) that performs user authentication, and U region of the image frames grouping bit information matching the received user information. Insert it in At this time, the reason for the insertion into the U area is that the color information area is insensitive to errors and light.

For example, when grouping n = 2 to include five image frames in each group (only for the first region of each of the five image frames), when the bit information to be inserted is 1, it corresponds to the center image frame of the group. The negative value is added to the U-domain so that the average value of the U-domain of the second image frame is smaller in the first image frame. A positive value is added to the U region (U-domain) of the fourth to fifth frames to sequentially increase the U region value from the first image frame to the fifth image frame. On the contrary, when the bit information to be inserted is 0, the U-domain is increased so that the average value of the U-domain of 1 to 2 image frames is increased around the third frame corresponding to the center frame of the group. Add a positive value. In addition, a negative value is added to the U region (U-domain) of the fourth to fifth frames to sequentially decrease the U region value from the first image frame to the fifth image frame.

The bit information inserting unit 140 inserts bit information into the U region and then converts the image frame converted into YUV into RGB.

The watermark embedding apparatus 100 may transmit an image frame to the watermark extraction apparatus 200 through a network, including a device that can be connected to a network such as a separate communication unit (omitted).

By using the watermark embedding apparatus and inserting multiple bits by dividing the area based on the pixel intensity value, the information protection rate can be significantly increased, and information loss due to compression, filtering, etc. can be minimized. In addition, unlike the conventional multi-bit insertion, it is possible to expect the effect of significantly increasing the invisibility and robustness by eliminating the difference of the peripheral edge value.

The watermark extracting apparatus 200 includes a frame arranging unit 210, a frame region dividing unit 220, a YUV converter 230, and a bit information extracting unit 240.

The watermark extraction apparatus 200 may include a device capable of connecting to a network such as a separate communication unit (omitted), and may receive a storage medium capable of receiving image information from the watermark embedding apparatus 100 or storing image information through the network. Image information can be received via (offline).

The frame arrangement unit 210 receives image information generated from the watermark embedding apparatus 100 and sequentially arranges image frames of the received image information in the order of input. Here, it is assumed that the image information received by the frame arrangement unit 210 is in the form of RGB.

The frame arranging unit 210 divides the arranged image frames into groups of 2n + 1 (where n = 1, 2, and 3 are natural numbers). In this case, the number of image frames in the group to which the frame arranging unit 210 is assigned is the same as the number n used by the watermark embedding apparatus 100 to allocate the group.

The frame region dividing unit 220 sequentially receives the image frames in group units from the frame arranging unit 210 and measures the intensity value I of the pixels using RGB of each image frame in the received group. The image frame is divided into four regions according to the histogram distribution of the intensity values I of the measured pixels. In this case, the number of divided regions is equal to the number of divided regions by the watermark embedding apparatus 100.

The YUV converter 230 converts the RGB of the image frame received from the frame region divider 220 into YUV and provides the converted information to the bit information extractor 240.

The bit information extractor 240 extracts the bit information including the user information by using the U region value of the YUV for each region (eg, the first region) of the image frame included in the group. For example, the bit information extractor 240 includes five image frames in each group by grouping n = 2 matching the watermark embedding apparatus 100 (only for the first region of each of the five image frames). ), The U-domain average values of the 1st to 2nd image frames corresponding to the front of the 3rd image frame corresponding to the center frame of the group, and the U region of the 4th to 5th image frames (U- By comparing the average value, if the U area average value of the 1st and 2nd image frames is greater than the U area average value of the 4th and 5th image frames, 0 bit information is extracted.

The bit information extractor 240 generates user information by using the extracted bit information, and provides the generated user information and an image frame to the user through a display apparatus (omitted).

Meanwhile, the watermark embedding apparatus 100 and the watermark extracting apparatus 200 according to an embodiment of the present invention may be included in a transmitting end (omitted) and a receiving end 200 receiving image information, respectively. It may be included in various devices for protecting image information.

Through such a watermark insertion / extraction device, there is a great advantage of minimizing the change of the original image of the video content while preventing illegal copying and distribution of the video content.

3 and 4 are diagrams illustrating a frame region divided by a frame region divider of each of a watermark embedding apparatus and a watermark extracting apparatus according to an exemplary embodiment of the present invention.

As shown in FIG. 3, the frame region dividers 120 and 220 according to an exemplary embodiment measure intensity values I of pixels using RGB of an image frame, and measure the intensity values I of the pixels. Segment the region through the histogram distribution for. For example, if an image frame having a size of 16 × 16 is divided into four regions, each region includes 64 pixels. First, the intensity values of all the pixels in the frame are calculated. The computed pixel intensity has a value in the range of 0-255. For example, suppose that the pixels in a frame have evenly distributed pixel intensities, that is, if 256 pixels have only one of 0 to 255 intensities, 64 pixels with intensities of 255 to 192 They belong to area 1. Similarly, 64 pixels having 191 to 128 intensity values belong to region 2, pixels of 127 to 64 intensity regions to region 3, and 64 pixels having intensity of 63 to 0 belong to region 4. The stronger the pixel intensity, the brighter the area. Here, the pixel intensity value I is equal to Equation 1 above.

In addition, as shown in FIG. 4, the frame region dividers 120 and 220 divide one image frame into four regions based on the pixel intensity.

Meanwhile, the watermark embedding apparatus and the watermark extracting apparatus according to an embodiment of the present invention divide one image frame into four regions, but may divide bit into N (N is a natural number) regions and insert bit information. . However, when dividing into five or more areas, the interference problem between adjacent areas should be considered.

In addition, when dividing one image frame into regions of N (N is a natural number), the watermark embedding apparatus and the watermark extracting apparatus set the same number of divided regions. To this end, the bit information inserting unit 140 of the watermark embedding apparatus 100 may include the number of divided regions in the image frame.

Next, a watermark embedding method and a watermark extracting method of a watermark embedding / extraction apparatus according to an embodiment of the present invention will be described with reference to FIGS. 5 and 6.

5 is a flowchart illustrating a watermark embedding method of a watermark embedding apparatus according to an embodiment of the present invention.

As shown in FIG. 5, in the watermark embedding method of the watermark embedding apparatus 100 according to an exemplary embodiment of the present invention, the frame arranging unit 110 illustrated in FIG. The video information is received from a device that provides video information such as a storage unit (omitted) for storing the information.

The frame arranging unit 110 arranges the image frames of the received image information in the order of input (S100), and divides the frame regions into groups of 5 (here, n = 2) for the arranged image frames. Provided in the installment 120 (S102).

The frame region dividing unit 120 measures intensity values I of pixels using RGB of each image frame in the group (S104), and uses a histogram distribution of the measured pixel intensity values I. The area is divided (S106). In each area where the user's information is placed, the total pixels of the frame are equally divided and allocated. In other words, when a frame is divided into N regions, pixels 1 having pixel intensity values of higher (1 / N)% are allocated to region 1. In area 2, the pixels having the next highest pixel intensity values are allocated to the last area N, and the remaining pixels having the lowest (1 / N)% intensity values are allocated. In other words, the pixel distribution is calculated according to the pixel intensity value, and an even number of pixels are allocated to each region to be divided.

The bit information insertion unit 140 provides information about the area of each image frame in the group.

The YUV converter 130 converts the RGB of the image frame received from the frame region divider 120 into YUV and provides the converted YUV to the bit information inserter 140 (S108).

The bit information inserting unit 140 receives user information of an authenticated user from an upper end such as an authentication unit (not shown) that performs user authentication, and allocates bit information matching the received user information to grouped image frames. The U area value of the area (eg, the first area) is changed and inserted (S110).

The bit information inserting unit 140 inserts the bit information by changing the U region value for the same region of the grouped image frames, and then, the watermark extracting apparatus 200 through a device connectable to a network such as a communication unit (omitted). In step S112, an image frame is transmitted.

Such a watermark embedding method has an advantage of inserting multi-bits on the basis of the pixel intensity value, thereby significantly increasing the information protection rate. In addition, it is possible to minimize information loss due to compression and filtering, and unlike the conventional multi-bit insertion, it is possible to expect the effect of significantly increasing the invisibility and robustness by eliminating the difference in the peripheral edge value.

6 is a flowchart illustrating a watermark extraction method of a watermark extraction apparatus according to an embodiment of the present invention.

As shown in FIG. 6, in the watermark extraction method of the watermark extraction apparatus 200 according to an exemplary embodiment, the frame arrangement unit receives image information generated from the watermark embedding apparatus 100.

The frame arranging unit 210 arranges the image frames of the image information received from the watermark embedding apparatus 100 in the order of input (S200), and 5 for the arranged image frames (assuming n = 2). The image is divided into groups and provided to the frame region dividing unit 220 (S202).

The frame region dividing unit 220 receives an image frame in unit units from the frame arranging unit 210 and divides the frame region for each image frame in the received group (S204).

The YUV converter 230 converts the RGB of the image frame received from the frame region divider 220 into YUV and provides the converted YUV to the bit information extractor 240 (S206).

The bit information extractor 240 extracts the bit information including the user information by comparing the U region value among the YUV-converted information of each region (eg, the first region) of the image frame included in the group ( S208).

The bit information extracting unit 240 generates user information using the extracted bit information (S210), and provides the generated user information and an image frame to the user through a display device (omitted) (S212).

The embodiments of the present invention described above are not only implemented through the apparatus and the method, but may also be implemented through a program for realizing a function corresponding to the configuration of the embodiment of the present invention or a recording medium on which the program is recorded. Implementation may be easily implemented by those skilled in the art from the description of the above-described embodiments.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements of those skilled in the art using the basic concepts of the present invention defined in the following claims are also provided. It belongs to the scope of rights.

1 is a diagram illustrating a segmentation of an image image by using a spatial method of inserting a multi-bit according to the prior art.

2 is a block diagram illustrating a watermarking system according to an exemplary embodiment of the present invention.

3 and 4 are diagrams illustrating a frame region divided by a frame region divider of each of a watermark embedding apparatus and a watermark extracting apparatus according to an exemplary embodiment of the present invention.

5 is a flowchart illustrating a watermark embedding method of a watermark embedding apparatus according to an embodiment of the present invention.

6 is a flowchart illustrating a watermark extraction method of a watermark extraction apparatus according to an embodiment of the present invention.

Claims (15)

In the watermark embedding apparatus for embedding a watermark in the image information, A frame arrangement unit for arranging the received image frames in an input order and generating a first group such that the arranged image frames are included in a set number; A frame region dividing unit dividing each image frame into two or more regions based on intensity values of pixels included in each image frame in the first group; And A bit information insertion unit for inserting bit information matching user information into a designated area among the divided areas Watermark embedding apparatus comprising a. The method of claim 1, The bit information insertion unit, And inserting the bit information by changing a color component of the designated area in a previous or subsequent image frame based on the reference image frame in the first group. The method of claim 2, The bit information insertion unit, And inserting the bit information by increasing or decreasing the color components by a predetermined number. The method according to any one of claims 1 to 3, And a conversion unit for converting RGB (Red, Green, Blue) components included in each image frame in the first group into U region values on YUV. The bit information insertion unit, And inserting the bit information by changing the color component of a U area value for the designated area. The method of claim 4, wherein The bit information insertion unit, A watermark embedding apparatus for receiving information of an authenticated user from an authentication unit for authenticating a user, and inserting the bit information matching the received user information. A watermark extraction apparatus for extracting watermark by receiving image information embedded with watermark, A frame arrangement unit for arranging image frames in an input order and generating a first group such that the arranged image frames are included in a set number; A frame region dividing unit which divides each image frame into two or more regions based on intensity values of pixels included in each image frame in the first group; And A bit information extracting unit for extracting bit information from a designated area among the divided areas Watermark extraction apparatus comprising a. The method of claim 6, The bit information extraction unit, And extracting bit information by comparing color components of a specified region of a previous image frame and a subsequent image frame based on the reference image frame in the first group. The method according to claim 6 or 7, And a conversion unit for converting RGB (Red, Green, Blue) components included in each image frame in the first group into U region values on YUV. The bit information extraction unit, And extracting the bit information by comparing a U area average value of the designated area of previous and subsequent image frames based on the reference image frame in the first group. In the watermark embedding method of the apparatus for embedding the watermark in the image information, Arranging image frames in an input order to generate a first group such that the image frames are included in a set number; Dividing each image frame into two or more regions based on pixel intensity values of each image frame in the first group; And Inserting bit information of user information into a designated region of each divided region of the image frame; Watermark embedding method comprising a. The method of claim 9, Inserting the bit information, And inserting the bit information by changing a color component of the divided region. The method of claim 10, Watermark embedding method further comprising the step of receiving the user information from the authentication unit for authenticating the user. The method according to any one of claims 9 to 11, Converting RGB (Red, Green, Blue) components included in each image frame in the first group into U-region values on YUV, Inserting the bit information, A watermark embedding method of inserting the bit information by changing the color component of a U region value for a designated region of each image frame in the first group A watermark extraction method of an apparatus for extracting watermark by receiving image information embedded with watermark, Arranging image frames in an input order to generate a first group such that the image frames are included in a set number; Dividing each image frame into two or more regions based on pixel intensity values of each image frame in the first group; And Extracting bit information from a designated area among the divided areas Watermark extraction method comprising a. The method of claim 13, Extracting the bit information, And extracting the bit information by comparing color components of a specified region of a previous image frame and a subsequent image frame based on the reference image frame in the first group. The method according to claim 13 or 14, Converting RGB (Red, Green, Blue) components included in each image frame in the first group into U-region values on YUV, Extracting the bit information, And extracting bit information from the U region average value of previous image frames and the U region average value of subsequent image frames based on the reference image frame in the first group with respect to the first region.

Images