KR102036077B1 - Method for processing images into which visible watermark is embedded, and image processing apparatus thereof - Google Patents

Abstract

A method of processing an image into which a visible watermark is embedded is disclosed. The method of processing an image includes: a process of storing image data of a predetermined region in which a visible watermark is embedded in an image; a process of obtaining a plurality of conversion luminance values corresponding to each of a plurality of pixels included in a unit block of the image; and a process of obtaining a plurality of quantization values corresponding to the plurality of conversion luminance values by quantizing the plurality of conversion luminance values. Here, the plurality of quantization values may include some of the image data of the predetermined region. It is possible to recover the image with the visible watermark efficiently and to minimize the quality deterioration of the recovered image.

Description

TECHNICAL FIELD OF THE INVENTION An image processing apparatus and a method for processing an image into which a visible watermark is embedded TECHNICAL FIELD

The present invention relates to a method of processing an image into which a visible watermark is inserted and an image processing apparatus thereof, and to an image into which a visible watermark is inserted for effectively processing an image of a portion into which a visible watermark is inserted. A method of processing and an image processing apparatus thereof.

Recently, with the development of the compression technology of multimedia data, high-quality digital content is rapidly and widely distributed. Digital content is easy to copy and transfer, and if it is leaked without permission, it will cause irreparable damage to the copyright holder of the digital content.

In order to prevent copyright infringement on digital content, various copyright protection techniques have been studied.

As a technique for protecting copyright of digital content, digital object identifier (DOI), digital right management (DRM), watermarking, and the like are known.

DOI technology is a copyright tracking technology that assigns a unique number to digital content and converts it into a uniform resource locator (URL), thereby making it easy to access a corresponding location on the Internet.

DRM technology protects the rights and interests of digital content providers and prevents illegal copying. It is a technology that supports the creation, distribution, and management of digital content such as charging royalties and settlement services.

Watermarking technology is a technology of invisibly inserting or extracting copyright information into digital content. In general, watermarking can be classified according to the type of work, the degree of recognition, and the purpose of use. Basically, the watermark technology is invisible to the human eye when the watermark is inserted, robustness to extract the watermark even in various forms of deformation or attack, and the extracted watermark can make a clear claim. Clarity is required.

For example, several prior arts are known for a method of embedding and extracting watermarks in joint photographic coding experts group (JPEG) image data.

For example, Korean Patent Publication No. 10-0620537 (September 13, 2006) describes a method of inserting or extracting a watermark using quantization values obtained by Huffman decoding on a digital image compressed with JPEG. This is disclosed.

As another example, Korean Patent Publication No. 2013-0085264 (September 23, 2013) extracts a quantized DCT coefficient through a discrete cosine transform (DCT) and a quantization process on an original image, and extracts the extracted quantization. A method of generating an authentication code using a DCT coefficient and inserting the generated authentication code into the quantized DCT coefficient extracted in the lossy compression process is performed for watermarking. A general watermarking technique has invisible properties like the above-described watermarking technique, but a watermarking technique (visible watermarking) having visible properties is also used.

The visible watermark embedded in the video content can most clearly convey the copyright holder of the video content. However, when restoring the original image by removing the visible watermark from the image, the quality of the restored image is deteriorated due to the removed visible watermark.

The present invention has been made to solve the above-described problem, various embodiments according to the present invention proposes a method for efficiently reconstructing the image embedded with a visible watermark, while minimizing the quality degradation of the reconstructed image.

According to various embodiments of the present disclosure, a method of processing an image into which a visible watermark is inserted may include storing image data of a predetermined region of the image into which the visible watermark is inserted. Obtaining a plurality of conversion luminance values corresponding to each of the plurality of pixels included in the unit block, and obtaining a plurality of quantization values corresponding to the plurality of conversion luminance values by quantizing the plurality of conversion luminance values. can do. Here, the plurality of quantization values may include some of the image data for the predefined area.

According to various embodiments of the present disclosure, an image processing apparatus for processing an image into which a visible watermark is inserted may include a storage unit and the storage unit to store image data of a predetermined region in which the visible watermark is inserted. Obtain a plurality of converted luminance values corresponding to each of the plurality of pixels included in the unit block of the image, and obtain a plurality of quantized values corresponding to the plurality of converted luminance values by quantizing the plurality of converted luminance values. It may include a processor. Here, the plurality of quantization values may include some of the image data for the predefined area.

According to various embodiments of the present disclosure, the quality degradation of the reconstructed image may be minimized while efficiently reconstructing the image into which the visible watermark is inserted.

1 illustrates a process of inserting and restoring a visible watermark according to an embodiment of the present invention.
2 is a layout view of a visible watermark according to an embodiment of the present invention.
3 illustrates a process of embedding a watermark according to an embodiment of the present invention.
4 is a diagram illustrating a reconstruction process of an original image, according to an exemplary embodiment.
5 illustrates a process of converting raw data according to an embodiment of the present invention.
6 illustrates a conversion process of a JPEG file according to an embodiment of the present invention.
7 is a block diagram of an image processing apparatus according to an exemplary embodiment.
8 illustrates a luminance quantization table according to an embodiment of the present invention.
9 illustrates a plurality of converted luminance values according to an embodiment of the present invention.
10 illustrates a method of determining quantization values according to an embodiment of the present invention.
11 illustrates a process of converting row data according to another embodiment of the present invention.
12 is a block diagram of a detailed configuration of an image processing apparatus according to an exemplary embodiment.
13 is a flowchart illustrating an image processing method according to an exemplary embodiment.

Hereinafter, the operation principle of the preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings. In addition, in describing the exemplary embodiment of the present invention, when it is determined that a detailed description of a related known function or configuration may obscure the subject matter of the present disclosure, the detailed description thereof will be omitted. The terms used below are terms defined in consideration of functions in the present invention, which may vary according to intentions or customs of users or operators. Therefore, the definitions of the terms used should be interpreted based on the contents throughout the present specification and the corresponding functions.

1 illustrates a process of inserting and restoring a visible watermark according to an embodiment of the present invention.

The digital watermark may include a visible watermark and an invisible watermark. The digital workmark may be defined as a bit pattern inserted in the digital content so that copyright information on the digital content can be identified. In addition, the visible watermark may be defined as a watermark that includes copyright information but is difficult to identify with a human eye. The invisible watermark includes copyright information but may be defined as a watermark that can be visually identified. The visible watermark and the invisible watermark may include various information related to digital content (eg, uniform resource locator (URL) information about a source of digital content) in addition to copyright related information. Hereinafter, a method of protecting copyright of digital content using a visible watermark and an invisible watermark will be described in detail with reference to various drawings.

Referring to FIG. 1, when the original image 11 is distributed, a visible watermark 121 may be inserted to protect copyright of the original image 11. For example, the visible watermark 121 may be a quick response code. Without being limited thereto, various logos including information on the copyright holder may be defined as the visible watermark 121.

When the visible watermark 121 is inserted into the original image 11, it is necessary to save (or store) the image data of the predefined area of the original image 11 on which the visible watermark 121 is disposed. . The image data preservation method for the predefined area of the original image 11 will be described later in detail.

The image 12 having the visible watermark 121 inserted therein may be distributed through various routes. For example, the image 12 into which the visible watermark 121 is inserted may be distributed through the Internet or recorded on a nonvolatile recording medium such as a USB, CD, or floppy disk.

Although the image 12 into which the visible watermark 121 is inserted may be used as it is for users, the user water quality of experience (QoE) is lower than that of the original image 11 due to the visible watermark 121. Can be.

In this case, the user can access the server providing the original image 11 by scanning the QR code, for example, the visible watermark 121, and watching the image from which the QR code has been removed, that is, the restored image 13, has high user satisfaction. Can be obtained. In addition, the copyright holder can induce traffic to the owner's server or service, and can expect a legitimate profit from the copyright.

2 is a layout view of a visible watermark according to an embodiment of the present invention.

Referring to FIG. 2, the visible watermark 22 may be disposed or inserted in a predetermined area of the image 21.

The size of the image 21 may be defined by the width 212 and the height 211.

The position of the visible watermark 22 may be defined by predefined coordinates 221. For example, the coordinates included in the image 21 may be defined as x, which is a coordinate corresponding to the horizontal axis, and y, which is a coordinate corresponding to the vertical axis. In this case, the predefined coordinate 221 may be defined as (x, y).

The size of the visible watermark 22 may be defined as a width 222 and a height 223. Here, the size of the visible watermark 22 may be determined so that copyright information of the image 21 can be inserted and cause QoE degradation on the image 21.

The size of the visible watermark 22 may be determined in proportion to the size of the image 21. In detail, as the visible watermark 22 is inserted into the image 21, the image data of the region where the visible watermark 22 is inserted may be lost. Accordingly, the image data of the region where the visible watermark 22 is inserted is inserted into the quantized value during the quantization process of the image 21 data according to various embodiments of the present invention, which will be described later. Can be restored on the basis of

Here, since the image data of the region where the visible watermark 22 is inserted must be included in all the quantized values in the quantization process of the data included in the image 21, the size of the visible watermark 22 is set to the image 21. It can be determined in proportion to the size of).

3 and 4, a process of inserting a watermark into the original image and a process of reconstructing the original image from the watermarked image will be described in detail.

3 illustrates a process of embedding a watermark according to an embodiment of the present invention.

Referring to FIG. 3, the watermark embedding process 30 inserts an invisible watermark into the original image 31 to generate the original image 32, and the visible water in the entire area of the original image 32. The process of generating the original image `` 33 by inserting image data of the area where the mark is placed and the image of inserting a visible watermark in the predefined area of the original image '' 33 (hereinafter referred to as 'the image distributed' ') 34 may be included. Here, the distributed image 34 may be distributed through various distribution channels.

4 is a diagram illustrating a reconstruction process of an original image, according to an exemplary embodiment.

Referring to FIG. 4, the reconstruction process 40 of the original image is a process of acquiring the original image '' 42 having the visible watermark removed from the distribution image 34 using the visible watermark detector 41, the original image. Restoring the original image '43 based on the image data of the region where the visible watermark inserted in the entire region of the image '' 42 is disposed, the original image using the invisible watermark detector 44 The method may include extracting the invisible watermark from '43, confirming copyright 46, and restoring the original image 45.

5 and 6, in the process of converting raw data into a JPEG file, a method of inserting image data of a predetermined area into which a visible watermark is inserted into an original image, a JPEG file A method of inserting image data of a predetermined area into which a visible watermark is inserted in a process of converting the JPEG file into an original image will be described in detail.

5 illustrates a process of converting raw data according to an embodiment of the present invention.

The conversion process 50 of raw data 501 is based on a human visual system (HVS). Also, it is assumed that the original image, that is, the raw data 501 is made of RGB.

The raw data 501 is converted into a luminance (Y, luminace) value 503 (or signal or information) and a color difference (CbCr, crominance) value 504, 505 through a color conversion 502. . This luminance value represents the contrast, and by using a human visual characteristic that recognizes the contrast better than the color difference, the color difference signal may be compressed more than the contrast signal to reduce the overall file size.

Down sampling 506 is a process of down-sampling a luminance value and a chrominance value to perform compression. In detail, the down sampling 506 converts the raw data 501 into the YCbCr color gamut and then performs the down sampling 506 on the visually insensitive Cb and Cr areas. For example, the down sampling 506 may perform Y: Cb: Cr at 4: 4: 4, 4: 2: 2, 4: 1: 1, and the like. For example, if Y: Cb: Cr is 4: 1: 1, downsampling 506 may determine the YCbCr value for four pixels by four Y values corresponding to four pixels, one Cb value, and one Cr value. It can be defined as down sampling 506 to one.

Discrete cosine transform (DCT) 507 is a process of transforming an image signal on a time axis into a frequency axis. Through the discrete cosine transform 507, the low frequency band among the DCT coefficients may be divided into a small change in color and a high frequency band in a large change in color. Human vision is more sensitive to areas of small color change than areas of large color change. This feature results in more lossy compression for the high frequency band, where a discrete cosine transform 507 can be used.

Quantization / Image Data Insertion 508 is a process of quantizing the transform luminance values resulting from the discrete cosine transform 507. For example, the quantization / image data insertion 508 may insert image data of a predefined area into which a visible watermark is inserted in the process of performing quantization based on the luminance quantization table shown in FIG. 8. Inserting process. The quantization / image data insertion 508 will be described later with reference to FIGS. 7 to 10.

Entropy encoding 509 is a process of encoding a YCbCr signal that has undergone a discrete coin transformation 507 and quantization / image data insertion 508.

Specifically, the YCbCr signal component that has undergone the discrete coin transformation 507 and the quantization / image data insertion 508 may be divided into a DC coefficient portion and an AC coefficient portion. Here, after collecting the DC coefficient components, Huffman encoding is performed using the difference value, and the AC coefficient component performs Huffman encoding using the difference value in the zigzag direction. In this case, the Huffman table is used.

When entropy encoding 509 is performed, the final file is stored 510 as a JPEG file, and the Huffman table may be stored within the JPEG file.

6 illustrates a conversion process of a JPEG file according to an embodiment of the present invention.

The conversion process 50 of the row data 501 of FIG. 5 described above is a case where the row data 501 is given as an input, and the conversion process 60 of the JPEG file 601 of FIG. Given as input. Hereinafter, a detailed description of processes overlapping with the conversion process 50 of the row data 501 of FIG. 5 will be omitted for convenience of description.

Referring to FIG. 6, the conversion process 60 of the JPEG file 601 is performed when the JPEG file 601 is input, Huffman decoding 602, de-quantization 603, and inverse-discrete cosine. Inverse-DCT (604), up-sampling (605), YCbCr value acquisition process (606, 607, 608), downsampling (609), discrete cosine transform (610), quantization / image data Insert 611, entropy encoding 612, and JPEG file storage 613.

Here, Huffman decoding 602 is the inverse of the Huffman encoding process included in entropy encoding 509 of FIG. 5. In addition, inverse quantization 603 is the inverse of the quantization process involved in quantization / image data insertion 508 of FIG. 5. However, the inverse of the image data insertion process included in the quantization / image data insertion 508 of FIG. 5 may not be performed. In addition, the inverse discrete cosine transform 604 is the inverse of the discrete cosine transform 507 of FIG. 5. Also, up-sampling 605 is the inverse of down sampling 506 of FIG. 5. Further, the YCbCr value acquisition process (606, 607, 608), downsampling (609), discrete cosine transform (610), quantization / image data insertion (611), entropy encoding (612) and JPEG file storage (613) are shown in FIG. Corresponds to the YCbCr value acquisition process (503, 504, 505), downsampling (506), discrete cosine transform (507), quantization / image data insertion (508), entropy encoding (509), and JPEG file storage (510) of 5. It is a process of becoming. Therefore, since each process of the conversion process 60 of the JPEG file 601 can be clearly understood with reference to each process of the conversion process 50 of the row data 501 of FIG. 5, redundant description is omitted here. .

In the above, with reference to FIGS. 5 and 6, the conversion process 50 of the raw data 501 in which the raw data is converted into a JPEG file, and the conversion process 60 of the JPEG file 601 in which the JPEG file is converted into a JPEG file ) Is described in detail.

The image data of the predefined area in which the above-described visible watermark is inserted in the original image is performed by performing the quantization / image data insertion 508 of the conversion process 50 of the raw data 501 and the JPEG file 601. In the process of performing the quantization / image data insertion 611 of the conversion process 60 of the "

In this way, the image data of the predefined area in which the above-described visible watermark is inserted in the entire original image is used in the process of restoring the original image, thereby equaling the quality of the original image before the visible watermark is inserted. To the original level can be restored.

An image processing apparatus that performs the process of inserting the image data for the predefined area where the above-mentioned visible watermark is inserted will be described in detail with reference to FIGS. 7 to 10.

7 is a block diagram of an image processing apparatus according to an exemplary embodiment.

Referring to FIG. 7, the image processing apparatus 700 may include a storage 710 and a processor 720.

The storage unit 710 stores data. For example, the storage unit 710 may store raw data, JPEG data, and the like for the image.

The processor 720 generally controls the image processing apparatus 700.

In particular, the processor 720 may process an image in which a visible watermark is inserted.

In detail, the processor 720 may control the storage unit 710 to store image data of a predefined area in which a visible watermark is inserted in the image (or the original image). The stored image data may be used to quantize a plurality of transform luminance values, which will be described later.

The processor 720 may obtain a plurality of converted luminance values corresponding to each of the plurality of pixels included in the unit block of the image. This operation of the processor 720 may be performed in the discrete cosine transform 507 of FIG. 5, the downsampling 506, and the discrete cosine transform 507. Alternatively, the operation of the processor 720 may be performed by the discrete cosine transform 610 of FIG. 6, the downsampling 609, and the discrete cosine transform 610.

 Here, the plurality of converted luminance values may be defined as discrete cosine transformed values of the plurality of luminance values corresponding to each of the plurality of pixels. For example, the plurality of luminance values may be Y values of FIGS. 5 and 6 described above.

The processor 720 may quantize the plurality of transform luminance values to obtain a plurality of quantization values corresponding to the plurality of transform luminance values. This operation of the processor 720 may be performed at the quantization / image data insertion 508 of FIG. 5 described above. Alternatively, the operation of the processor 720 may be performed by the quantization / image data insertion 611 of FIG. 6 described above.

Here, the plurality of quantization values may include some of image data of a predefined area in which a visible watermark is inserted in the image. For example, each of the plurality of quantization values may include information about a unit bit included in a bit string corresponding to a part of image data for a predefined area. For example, when the image data of the predefined area in which the visible watermark is inserted in the image includes a bit string of '1100101', '1' is embedded in the first value of the plurality of quantization values. , '1' is inserted into the second value of the plurality of quantization values, '0' is inserted into the third value of the plurality of quantization values, '0' is inserted into the fourth value of the plurality of quantization values, and '1' is inserted into a fifth value of the quantization value of, '0' is inserted into a sixth value of the plurality of quantization values, and '1' is inserted into a seventh value of the plurality of quantization values.

The plurality of quantization values described above may be determined based on a plurality of predefined step sizes corresponding to each of the plurality of conversion luminance values. Hereinafter, a method of determining a plurality of quantization values will be described in detail with reference to FIGS. 8 and 9.

8 illustrates a luminance quantization table 80 that includes a plurality of predefined step sizes described above. Since the discrete cosine transform is performed on an 8 × 8 block based on the pixel, the luminance quantization table may also be configured as an 8 × 8 block. Here, the luminance quantization table 80 may be stored in various ways. In addition, the luminance quantization table 80 may be selectively used based on characteristics associated with an original image or a visible watermark of the image. For example, the luminance quantization table 80 may be selectively used according to characteristics such as luminance characteristics between a region into which a visible watermark is inserted and the rest of the original image, a size of a region into which the visible watermark is inserted, and the like.

Referring to FIG. 8, the first step size 81 of the luminance quantization table is 16, the second step size 82 is 6, and the third step size 83 is 6. Since the fourth step size 84 is overlapping content, description thereof is omitted. Here, the order of loading each step size may be set in a zigzag order such as reference numeral 81, reference numeral 82, reference numeral 83, and reference numeral 84.

For example, in FIG. 8, at least some of the step sizes of the area indicated by gray may be used for luminance quantization. In addition, in FIG. 8, step sizes other than the area indicated in gray may not be utilized for luminance quantization.

9 illustrates a plurality of converted luminance values according to an embodiment of the present invention.

Referring to FIG. 9, through a discrete cosine transform process, the first transform luminance value 91 is 45, the second transform luminance value 92 is 99, the third transform luminance value 93 is 99, and the fourth transform luminance. The value 94 is obtained as 99. Since the fourth converted luminance value 94 is overlapped, the description thereof is omitted. Here, the order of loading the converted luminance values may be set in a zigzag order as shown by reference numeral 91, reference numeral 92, reference numeral 93, and reference numeral 94.

The processor 720 may determine the plurality of quantization values based on the plurality of predefined step sizes 80 and the plurality of transform luminance values 90.

Since the first value 45 of the plurality of converted luminance values 90 is treated as an important value DC in luminance, the first value 91 of the plurality of converted luminance values 90 is quantized. In the meantime, it is preferable not to insert the image data for the predefined area in which the visible watermark is inserted. Hereinafter, it will be described that image data of a predetermined area in which a visible watermark is inserted in the image is inserted from the first value 92 of the plurality of converted luminance values 90.

First, a method of quantizing the first value 91 of the plurality of converted luminance values 90 will be described in detail.

The processor 720 may include a plurality of quantization values based on a first value having a smallest difference from 45 among multiples of the predetermined step size 81 and 16 corresponding to 45 that is one of the plurality of converted luminance values 90. You can decide to either.

For example, the processor 720 may be a multiple of the predefined step size 81 value 16 corresponding to 45 which is one 91 of the plurality of converted luminance values 90 (16, 32, 48, 64,...) In this case, 48 having the smallest difference from 45 which is one 91 of the plurality of converted luminance values 90 may be determined as the quantization value.

Hereinafter, a method of inserting bit string data into the plurality of quantized values while converting the plurality of converted luminance values 90 into the plurality of quantized values will be described in detail. Hereinafter, it is assumed that a bit string that is a part of image data for a predefined area in which a visible watermark is inserted in the image is '100'.

First, a method of inserting one-bit unit bit '1' into the quantization value while quantizing the second value 92 of the plurality of converted luminance values 90 will be described in detail.

The processor 720 is most different from one of the plurality of converted luminance values 90 among multiples of the predefined step size 82 corresponding to one of the plurality of converted luminance values 90. One of the second smallest value and the second smallest difference from one of the plurality of converted luminance values may be determined as one of the plurality of quantization values.

For example, the processor 720 may be a multiple of 6, 12, 18, 24, 30, 36 of the defined step size 82 6 corresponding to 99, which is one of the plurality of converted luminance values 90. , 42, 48, 54, 60, 66, 72, 78, 84, 90, 96, 102, ...), the second smallest difference from 99 and the second smallest value 96 from 99 May be determined as one of 99 quantization values.

Here, if the quotient of the result of dividing the first value or the second value by the predetermined step size is odd, the processor 720 includes unit bit information included in the first value or the second value is '1', and the first If the quotient of the result of dividing the value or the second value by the predefined step size is an even number, the unit bit information included in the first value or the second value may be determined as '0'.

Based on the above-described embodiment, since the quotient 17 of the result of dividing the first value 102 by the predefined step size 6 is an odd number, the unit bit information inserted into the first value 48 is '1', and the second Since the quotient 16 of the result of dividing the value 96 by the predefined step size 6 is an even number, the unit bit information included in the second value 96 is '0'.

Since the quantization value for 99, one 92 of the plurality of transform luminance values 90, must include unit bit '1', process 720 may determine the final quantization value as 102.

Next, a method of inserting one-bit unit bit '0' into the quantization value while quantizing the third value 93 99 among the plurality of converted luminance values 90 will be described in detail.

The processor 720 is most different from one of the plurality of converted luminance values 90 among multiples of the predefined step size 83 corresponding to one of the plurality of converted luminance values 90. One of the second smallest value and the second smallest difference from one of the plurality of converted luminance values may be determined as one of the plurality of quantization values.

For example, processor 720 may be a multiple of 6, 12, 18, 24, 30, 36 of predefined step size 83 6 corresponding to 99, which is one of plurality of converted luminance values 90. , 42, 48, 54, 60, 66, 72, 78, 84, 90, 96, 102, ...), the second smallest difference from 99 and the second smallest value 96 from 99 May be determined as one of 99 quantization values.

Here, if the quotient of the result of dividing the first value or the second value by the predetermined step size is odd, the processor 720 includes unit bit information included in the first value or the second value is '1', and the first If the quotient of the result of dividing the value or the second value by the predefined step size is an even number, the unit bit information included in the first value or the second value may be determined as '0'.

Based on the above-described embodiment, since the quotient 17 of the result of dividing the first value 102 by the predefined step size 6 is an odd number, the unit bit information inserted into the first value 102 is '1' and the second Since the quotient 16 of the result of dividing the value 96 by the predefined step size 6 is an even number, the unit bit information included in the second value 96 is '0'.

Since the quantization value for 99, one of the plurality of transform luminance values 90, must include unit bit '0', process 720 may determine the final quantization value as 96.

Next, a method of inserting one-bit unit bit '0' into the quantization value while quantizing the fourth value 94 99 of the plurality of converted luminance values 90 will be described in detail.

The processor 720 is most different from one of the plurality of converted luminance values 90 among multiples of the predefined step size 84 corresponding to one of the plurality of converted luminance values 90. One of the second smallest value and the second smallest difference from one of the plurality of converted luminance values may be determined as one of the plurality of quantization values.

For example, the processor 720 may be a multiple of the predefined step size 84 7 (7, 14, 21, 28, 35, 42) corresponding to 99, which is one of the plurality of converted luminance values 90 (94). , 56, 63, 70, 77, 84, 91, 98, 105, ...) quantization of one of the second value 105, the second smallest difference from 99 and 99, the smallest difference from 99 Can be determined by one of the values.

Here, if the quotient of the result of dividing the first value or the second value by the predetermined step size is odd, the processor 720 includes unit bit information included in the first value or the second value is '1', and the first If the quotient of the result of dividing the value or the second value by the predefined step size is an even number, the unit bit information included in the first value or the second value may be determined as '0'.

Based on the above-described embodiment, since the quotient 14 of the result of dividing the first value 98 by the predefined step size 7 is an even number, the unit bit information inserted into the first value 98 is '0', and the second Since the quotient 15 of the result of dividing the value 105 by the predefined step size 7 is odd, the unit bit information included in the second value 105 is '1'.

Since the quantization value for 99, one of the plurality of transform luminance values 90, must include unit bit '0', process 720 may determine the final quantization value as 98.

In the above-described exemplary embodiment of the present invention, the image data is inserted into the converted luminance value corresponding to the step size of the gray area shown in FIG. 8 among the plurality of converted luminance values 90 of FIG. 9. can do. In addition, image data may be inserted into a color difference value such as CbCr as well as the plurality of converted luminance values 90 of FIG. 9.

In the above-described embodiment of the present invention, the final quantization values '102', '96, '98' including the bit string '100' which is a part of the image data for the predefined area in which the visible watermark is inserted in the image are determined. The process of the described in detail. Referring to the above-described embodiment of the present invention, a visible watermark is inserted in the image by referring to the quantization values '102', '96, '98' and the luminance quantization table 90 in the receiving end obtaining the final quantization value. A bit string '100' which is a part of image data for a predefined area may be obtained. That is, according to various embodiments of the present disclosure, the high quality of the reconstructed image may be realized by effectively reconstructing the data of the image into which the visible watermark is inserted.

10 illustrates a method of determining quantization values. To insert '0' or '1' into the quantization value, select the second value closest to the transform luminance value among the multiples of the step size as the quantization value, but the second value closest to the nearest value. One is defined as containing '0' and the other is defined as containing '1'.

In the various examples of the present invention described above, the description is based on the JPEG compression method for convenience of description, but the present invention is not limited thereto. For example, various embodiments of the present invention may be applied to a video compression method based on a moving picture experts group (MPEG).

Hereinafter, an example in which various embodiments of the present invention described above are applied to MPEG will be described with reference to FIG. 11. Detailed descriptions of components overlapping with those of the exemplary embodiment of FIG. 5 will be omitted.

11 illustrates a process of converting row data according to another embodiment of the present invention.

Referring to FIG. 11, a video 1101 may be separated into an image 1103 and an audio 1104 by a demuxer 1102.

MPEG-1 / MPEG-2 uses the concept of group of pictures (GOP). The GOP consists of a frame type of an I frame (Intra frame), a P frame (Predicted frame) and a B frame (Bidirectional frame). The I frame is the most important frame that is a standard of image quality, the P frame is a frame constructed based on the information of the previous I frame, and the B frame is a frame composed based on the information of the I or P frame before and after.

The above-described original image 1103 of FIG. 11 may be composed of I frames, P frames, and B frames.

Here, although the above-described process of FIG. 5 has been described as being performed based on JPEG, the process of FIG. 11 corresponding to the above-described processes 502 to 509 of FIG. 1105 to 1112 may be performed on the I frame 1113.

4 and 11, an I frame 1113 generated through steps 1105 to 1112 of FIG. 11 is generated as an original image ″ 1114 with a visible watermark embedded therein, and is combined with an audio 1115. The video may be inputted to the Muxer 1116 and reproduced as a video 1117.

Although the image processing method for the I frame having high importance has been described in the above-described embodiment of FIG. 11, the image processing method described above may be appropriately applied to the P frame and the B frame.

In addition, although the above-described image processing method has been described as being applied to MPEG-1 / MPEG-2, the present invention is not limited thereto, and the same may be applied to MPEG-4 at the same level of technology.

On the other hand, in the video 1101 as a whole, when the amount of motion change of an object included in the video 1101 is large, the frequency of I frames in the video 1101 increases. In this case, since the frequency of the I frames increases, data about the original video corresponding to the visible watermark region can be sufficiently inserted into the I frames.

On the contrary, while the amount of change of the motion of the object included in the visible watermark area of the video 1101 is greater than or equal to the predefined change amount, the amount of change of the motion of the object included in the non-visible watermark area is less than or equal to the predetermined change amount. In this case, data about the original video corresponding to the visible watermark region is increased. In this case, in order to store data about the increased original video, the processor 720 may increase the number (or ratio) of I frames included in the video 1101. That is, the number (or frequency) of I frames included in the video 1101 may be proportional to the amount of change in movement of an object included in the visible watermark. Here, the motion change amount may be defined as a change amount of a motion vector or a change amount of motion estimation.

12 is a block diagram of a detailed configuration of an image processing apparatus according to an exemplary embodiment.

Referring to FIG. 12, the electronic device 1200 includes a communication unit 1210, a storage unit 1220, and a processor 1230.

The communication unit 1210 performs communication. The communication unit 1210 may communicate with an external device through various communication methods such as BT (BlueTooth), WI-FI (Wireless Fidelity), Zigbee, IR (Infrared), NFC (Near Field Communication), and the like.

The storage unit 1220 may store an operating system (O / S) software module for driving the electronic device 1200, data for configuring various UI screens provided in the display area, and the like. In addition, the storage unit 1220 may read and write.

The processor 1230 generally controls the operation of the image processing apparatus 1200 by using various programs stored in the storage 1220.

In detail, the processor 1230 may use the RAM 1231, the ROM 1232, the main CPU 1233, the graphic processor 1234, the first to n interfaces 1235-1 to 1235-n, and the bus 1236. Include.

The RAM 1231, the ROM 1232, the main CPU 1233, the graphic processor 1234, the first to nth interfaces 1235-1 to 1235-n, and the like may be connected to each other through the bus 1236.

The first to n interfaces 1235-1 to 1235-n are connected to the various components described above. One of the interfaces may be a network interface connected to an external device via a network.

The main CPU 1233 accesses the storage unit 1220 and performs booting using an operating system stored in the storage unit 1220. The controller 1220 performs various operations by using various programs, content, data, and the like stored in the storage 1220.

The ROM 1232 stores a command set for system booting. When the turn-on command is input and power is supplied, the main CPU 1233 copies the O / S stored in the storage unit 1220 to the RAM 1231 according to the command stored in the ROM 1232, and executes O / S. Boot the system. When the booting is completed, the main CPU 1233 copies various stored application programs to the RAM 1231, and executes the application programs copied to the RAM 1231 to perform various operations.

The graphic processor 1234 generates a screen including various objects such as an icon, an image, and a text by using the calculator and the renderer.

13 is a flowchart illustrating an image processing method according to an exemplary embodiment.

Referring to FIG. 13, in the method of processing an image into which a visible watermark is inserted, a process of storing image data of a predefined area in which a visible watermark is inserted in operation 1310 and a plurality of units included in a unit block of the image Acquiring a plurality of transform luminance values corresponding to each of the pixels 1320, and obtaining a plurality of quantization values corresponding to the plurality of transform luminance values by quantizing the plurality of transform luminance values (1330). have.

Here, the plurality of quantization values may include some of the image data for the predefined area.

Each of the plurality of quantization values described above may include information about a unit bit included in a bit string corresponding to a part of the image data for the predefined area.

The plurality of transform luminance values described above may be discrete cosine transform (DCT) of a plurality of luminance values corresponding to each of the plurality of pixels.

In addition, the plurality of quantization values described above may be determined based on a plurality of predefined step sizes corresponding to each of the plurality of conversion luminance values.

Here, one of the plurality of quantization values is a first value and a plurality of conversion luminance values having a smallest difference from one of the plurality of conversion luminance values among multiples of a predefined step size corresponding to one of the plurality of conversion luminance values. May be one of the second smaller second values.

In addition, if the quotient of the result of dividing the first value or the second value by the predetermined step size is odd, the unit bit information included in the first value or the second value is If it is '1' and the quotient of the result of dividing the first value or the second value by the predefined step size is an even number, the unit bit information included in the first value or the second value may be '0'.

The image into which the above-described visible watermark is inserted may be an image into which an invisible digital watermark is inserted.

Meanwhile, the image processing method according to various exemplary embodiments of the present disclosure described above may be implemented as computer executable program code and executed by a processor in a state of being stored in various non-transitory computer readable mediums. Or in devices.

For example, storing image data of a predetermined area in which a visible watermark is inserted in the image, acquiring a plurality of converted luminance values corresponding to each of the plurality of pixels included in the unit block of the image, and a plurality of A non-transitory computer readable medium may be provided in which a program for performing a process of obtaining a plurality of quantization values corresponding to the plurality of transform luminance values by quantizing the transform luminance values.

A non-transitory readable medium refers to a medium that stores data semi-permanently and is read by a device, not a medium that stores data for a short time such as a register, a cache, or a memory. Specifically, the various applications or programs described above may be stored and provided in a non-transitory readable medium such as a CD, a DVD, a hard disk, a Blu-ray disk, a USB, a memory card, a ROM, or the like.

In addition, while the above has been shown and described with respect to preferred embodiments of the present disclosure, the present disclosure is not limited to the specific embodiments described above, the technical field to which the present invention belongs without departing from the spirit of the present disclosure claimed in the claims. Of course, various modifications can be made by those skilled in the art, and these modifications should not be individually understood from the technical spirit or the prospect of the present disclosure.

Image Processing Unit: 700
Communications Department: 1110
Storage: 710, 1120
Processor: 720, 1230

Claims (18)

In the method for processing an image in which a visible watermark is inserted,
Storing image data of a predetermined area of the image into which the visible watermark is inserted;
Obtaining a plurality of converted luminance values corresponding to each of the plurality of pixels included in the unit block of the image;
Quantizing the plurality of transform luminance values to obtain a plurality of quantized values corresponding to the plurality of transform luminance values;
The plurality of quantization values,
And a portion of the image data for the predefined area.
The method of claim 1,
Each of the plurality of quantization values,
And information about a unit bit included in a bit string corresponding to the portion of the image data for the predefined area.
The method of claim 1,
The plurality of converted luminance values are,
And a plurality of luminance values corresponding to each of the plurality of pixels are discrete cosine transform (DCT) values.
The method of claim 1,
The plurality of quantization values,
And determined based on a plurality of predefined step sizes corresponding to each of the plurality of converted luminance values.
The method of claim 4, wherein
One of the plurality of quantization values,
A difference between the first value and the one of the plurality of converted luminance values having the smallest difference from one of the plurality of converted luminance values among multiples of a predefined step size corresponding to one of the plurality of converted luminance values And one of the second smallest second values.
The method of claim 5,
The first value or the second value is,
If the quotient of the result of dividing the first value or the second value by the predefined step size is odd, the unit bit information included in the first value or the second value is '1',
And if the quotient of the result of dividing the first value or the second value by the predefined step size is an even number, the unit bit information included in the first value or the second value is '0'.
The method of claim 1,
And an invisible digital watermark is inserted in the image into which the visible watermark is inserted.
In claim 1,
The visible watermark is a QR code (quick response code), the image processing method.
The method of claim 1,
The video,
An image processing method, which is an I frame of a joint photographic coding experts group (JPEG) picture or a moving picture experts group (MPEG) picture.
An image processing apparatus for processing an image into which a visible watermark is inserted, the image processing apparatus comprising:
Storage unit; And
Controlling the storage unit to store image data of a predetermined area of the image into which the visible watermark is inserted;
Acquire a plurality of converted luminance values corresponding to each of the plurality of pixels included in the unit block of the image,
And a processor configured to quantize the plurality of converted luminance values to obtain a plurality of quantized values corresponding to the plurality of converted luminance values.
The plurality of quantization values,
And a portion of the image data for the predefined area.
The method of claim 10,
Each of the plurality of quantization values,
And information about a unit bit included in a bit string corresponding to the part of the image data for the predefined area.
The method of claim 10,
The plurality of converted luminance values are,
And a plurality of luminance values corresponding to each of the plurality of pixels are discrete cosine transform (DCT) values.
The method of claim 10,
The plurality of quantization values,
And are determined based on a plurality of predefined step sizes corresponding to each of the plurality of converted luminance values.
The method of claim 13,
One of the plurality of quantization values,
A difference between the first value and the one of the plurality of converted luminance values having the smallest difference from one of the plurality of converted luminance values among multiples of a predefined step size corresponding to one of the plurality of converted luminance values And one of the second smallest second values.
The method of claim 14,
The first value or the second value is,
If the quotient of the result of dividing the first value or the second value by the predefined step size is odd, the unit bit information included in the first value or the second value is '1',
And if the quotient of the result of dividing the first value or the second value by the predefined step size is an even number, the unit bit information included in the first value or the second value is '0'.
The method of claim 10,
And an invisible digital watermark is inserted in the image into which the visible watermark is inserted.
In claim 10,
The visible watermark is a QR code (quick response code), the image processing device.
The method of claim 10,
The video,
An image processing apparatus, which is an I frame of a joint photographic coding experts group (JPEG) picture or a moving picture experts group (MPEG) picture.

Images