JP2003348553A - Digital watermark-embedding apparatus, method thereof, and recording medium having recorded therein program for performing the method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology which makes a degradation in image quality compatible with compression resistance by adjusting embedment intensity. <P>SOLUTION: An input image is divided into local regions, each of which is a macro block unit. Embedment intensity is determined for each of the local regions using calculated characteristic amount and a compression bit rate. The determined embedment intensity is used to insert digital watermarks into each of the local regions, and the input image is encoded, thereby outputting a compressed bit stream that has the digital watermarks embedded therein. Since a characteristics amount is calculated for each of small regions of the input image, the embedment intensity can be adjusted for each region and the degradation in image quality can be suppressed. A correspondence table corresponding to a predetermined compressed bit rate is used to adjust the embedment intensity, thereby determining the embedment intensity while taking into consideration the image quality of a digital watermark-inserted image subjected to image encoding. <P>COPYRIGHT: (C)2004,JPO

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital watermark embedding device, a method thereof, and a medium storing a program for executing the method, and more particularly to a method of adjusting the strength of embedding a digital watermark in a digital image. .

[0002]

2. Description of the Related Art In recent years, since digital contents such as digitized audio and video data can be easily duplicated exactly as the original, copyright protection of the contents has become an important issue. As one of the solutions,
"Digital watermark" is used.

[0003] Digital watermarking is a technique for embedding data in video data so that humans cannot perceive image quality degradation, and detecting the embedded data from the embedded image data.

Further, even from an image on which image processing such as image compression and noise addition has been performed, it is necessary to devise a digital watermark embedded in the image data so that it can be correctly detected. In the present specification, when such image processing is performed, the degree to which the digital watermark withstands detection is referred to as resistance to image processing.

[0005] There is a trade-off between the image quality degradation and the resistance to image processing, and the balance between the two can be achieved by adjusting the value of the embedding strength.

For example, when the value of the embedding strength is increased, the image quality of the image in which the digital watermark is embedded becomes larger, but the resistance to the image processing is improved. Conversely, when the value of the embedding strength is reduced, image quality degradation can be suppressed, but the durability decreases.

An example of a conventional digital watermark embedding method that automatically adjusts such a trade-off relationship is disclosed in Japanese Patent Application Laid-Open No. H11-163,878. Hereinafter, a conventional digital watermark embedding method described in this document will be briefly described.

FIG. 4 is a block diagram of a conventional digital watermark insertion system. In FIG. 4, the category classification unit 40
1 calculates a feature amount of the input image, obtains a category to which the image belongs from the obtained feature amount, and outputs a category index to the storage device 402.

The storage device 402 has, for each category index, a digital watermark characteristic table in which the correspondence between the digital watermark strength, the image quality deterioration degree, and the resistance evaluation value against attack (image processing) is described. The digital watermark characteristic table to be used is selected according to the category index to be used.

[0010] The digital watermark strength calculation unit 403 outputs the digital watermark strength to the storage device 402, and uses the image quality deterioration degree and the resistance evaluation value output from the storage device 402, and constraint information input by the user, and The optimum digital watermark strength is calculated and output to the digital watermark insertion unit 404.

A digital watermark insertion unit 404 converts the embedded data into a digital watermark, and
The digital watermark is inserted into the input image at the optimum digital watermark strength input from Step 3 and the digital watermark inserted image is output.

(1) In the prior art, the activity of the entire image (mean square value of the AC frequency component) is used as the feature amount of the input image.

However, the input image is not necessarily a uniform image as a whole, and in most cases there are complex portions with many changes and monotonous portions with little change.

If the embedding strength of the digital watermark is too strong for the image, the deterioration of the image quality due to the digital watermark becomes conspicuous. In particular, in a portion where the change is small, the deterioration of the image quality becomes more remarkable.

Therefore, according to the prior art, the embedding strength of a digital watermark is often inappropriate in an input image in some cases, and there is a problem that image quality degradation is inevitable.

(2) The image quality deterioration degree in the prior art is a numerical value calculated from the digital watermark inserted image immediately after the digital watermark is inserted and the input image.

However, when video data is actually stored and stored on a recording medium such as a hard disk and then transmitted and distributed over a network or the like, in order to reduce the amount of data, the image constituting the video data must be a moving image. if there is,
For example, in the case of a still image such as an MPEG method, for example, JP
Encoding is performed according to the EG method or the like.

For this reason, the digital watermark inserted image before encoding and the digital watermark inserted image after encoding are
Generally different.

For example, in a digital watermark inserted image compressed at a low compression bit rate, block distortion due to compression appears remarkably.
The image quality is greatly deteriorated.

That is, in the prior art, since no change in image quality before and after compression is taken into account, a digital watermark cannot be embedded in an input image with a preferable embedding strength. There was a problem that the relationship was likely to be inappropriate.

[Patent Document 1] JP-A-11-346302 (pages 9-11, FIG. 1)

[Patent Document 2] JP-A-2001-352441 (page 3-4, FIG. 1)

[0021]

SUMMARY OF THE INVENTION A first object of the present invention is to provide a technique capable of suppressing image quality deterioration.

A second object of the present invention is to provide a technique for embedding a digital watermark with an appropriate embedding strength in an input image even when performing compression.

[0023]

According to a first aspect of the present invention, there is provided a digital watermark embedding apparatus for calculating a feature amount of a digital image, a feature amount calculated by the feature amount calculation unit, and a feature amount used in encoding. An embedding strength determining unit for obtaining an embedding strength using a compression bit rate to be embedded, a digital watermark inserting unit for embedding a digital watermark in a digital image using the embedding strength determined by the embedding strength determining unit, An image encoding unit that encodes the embedded digital image at a compression bit rate and outputs a compressed bit stream.

According to this configuration, the digital watermark strength can be obtained based on the compression bit rate used for encoding and the feature amount of the input image. That is, the embedding strength of the digital watermark is obtained in consideration of the image quality of the digital watermark inserted image after image encoding, and the digital watermark can be embedded in the digital image by reflecting the degree of compression.

A second invention is an invention according to the first invention, wherein the feature value calculating section calculates a feature value for each partial region constituting a part of the digital image.

In this configuration, by calculating the feature amount for each partial region of the input image, the embedding strength can be adjusted for each partial region, and image quality degradation can be suppressed. That is, since the digital watermark is embedded for each partial region according to the state of the partial region by the embedding strength according to the feature amount, it is possible to suppress the image quality deterioration due to the embedding of the digital watermark.

[0027] A third invention is an invention according to the first invention, wherein the embedding strength judging unit includes a feature amount calculated by the feature amount calculating unit and a compression bit rate used in encoding of the image encoding unit. And a correspondence table of one or both of the above and the embedding strength of the digital watermark is used.

By using the correspondence table in this configuration, the embedding strength can be adjusted by reflecting the compression bit rate. That is, the digital watermark can be embedded in the digital image by the embedding strength in consideration of the image quality of the digital watermark inserted image after the image encoding.

[0029]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a digital watermark embedding device according to one embodiment of the present invention.

As shown in FIG. 1, the digital watermark embedding apparatus includes a feature amount calculating unit 101, an embedding strength determining unit 102, a digital watermark inserting unit 103, and an image encoding unit 1.
04. Here, the image encoding unit 104 according to the present embodiment corresponds to a processing unit that performs predetermined signal processing on a digital image in which a digital watermark is embedded.

The details of the digital watermark embedding process will be described below with reference to FIGS. FIG.
5 is a flowchart of the digital watermark embedding device.

First, the characteristic amount calculation unit 101 divides an input digital image into a plurality of partial region images in a predetermined pixel unit, and calculates a characteristic amount in partial region units (step S201).

The partial area in the present embodiment is a macro block having a size of 16 × 16 pixels. Further, the feature amount for each partial region is an average value of the luminance component.

Here, as described above, the partial area is a processing unit corresponding to image coding, such as one macro block, a plurality of macro blocks constituting one line, or an object unit. It may be changed as follows.

When a plurality of macroblocks constituting one line are used as partial areas, the amount of calculation can be reduced as compared with a case where one macroblock is used as a partial area. Therefore, the present invention is easily applied when there is not enough system resources.

In addition to the average value of the luminance component,
The following sum of the absolute differences of the horizontal and vertical adjacent pixels may be used. Here, the partial area is 16 × 1
It has a size of 6 pixels, and the luminance data of each pixel is d (i,
j) When (0 ≦ i ≦ 15, 0 ≦ j ≦ 15), the total sum Sh of absolute differences between horizontally adjacent pixels can be defined by the following equation.

[0037]

(Equation 1)

The sum Sv of absolute differences between adjacent pixels in the vertical direction can be defined by the following equation.

[0039]

(Equation 2)

Further, when a characteristic amount suitable for human visual characteristics is calculated and executed, there is an effect that image quality deterioration is hardly perceived and the embedding strength can be increased. Of course, the above-described feature amounts are merely examples, and the present invention includes a case in which another feature amount is used.

Next, the processing of the embedding strength determination unit 102 (step S202) will be described with reference to FIG. Table 1 shown in FIG. 3A and Table 2 shown in FIG. 3B are correspondence tables of the embedding strength with respect to the feature amount and the compression bit rate, respectively.

When the table 1 is used, the characteristic amount calculated by the characteristic amount calculating section 101 and a predetermined threshold value Y
1, Y2 (Y2 <Y1) are compared. Then, “strength 1” is obtained based on the range of the feature amount. For example, when the value of the feature amount is larger than the threshold value Y1,
“3” is output as “strength 1”.

When Table 2 is used, the compression bit rate used by the image encoding unit 104 is compared with predetermined thresholds B1 and B2 (B2 <B1). Then, based on the compression bit rate range, “strength 2”
Is required. For example, when the value of the compression bit rate is smaller than the threshold value B2, “3” is set as “strength 2”.
Is output.

Further, the embedding strength judgment unit 102
The “strength 1” and the “strength 2” are added, and the sum is output to the digital watermark insertion unit 103 as the embedding strength.
In the above example, the embedding strength is 3 + 3 = 6.

In the above description, the output embedding strength is the sum of "strength 1" and "strength 2", but the present invention is not limited to this. The output embedding strength is
For example, the product of “strength 1” and “strength 2” may be used.

Next, the digital watermark insertion unit 103 embeds a digital watermark in an image of a partial area (here, a macroblock) using the embedding strength obtained by the embedding strength determination unit 102, and obtains an image of the embedded partial area. Is output to the image encoding unit 104 (step S203).

By doing so, the characteristic amount calculation unit 101,
The respective processes of the digital watermark insertion unit 103 and the image encoding unit 104 proceed in parallel in macroblock units, so that the waiting time is small and the efficiency is high.

Next, the image encoding unit 104 encodes the image of the partial area into which the digital watermark has been inserted, according to a predetermined compression bit rate, and outputs a compressed bit stream (step S204). .

As described above, according to the digital watermark embedding device according to the present embodiment, a digital image is divided into a plurality of partial regions in a predetermined pixel unit, and the feature amount for each partial region is calculated. Then, the embedding strength is determined using a correspondence table corresponding to the feature amount and a predetermined compression bit rate.

As a result, by calculating the feature amount for each small area of the input image, the embedding strength for each area can be adjusted, and image quality deterioration can be suppressed. In addition, by adjusting the embedding strength according to the compression bit rate,
The embedding strength can be determined in consideration of the image quality of the digital watermark inserted image after the image encoding.

Typically, each function realized by the digital watermark embedding device according to the present embodiment is implemented by a storage device (ROM, RAM, hard disk, etc.) storing predetermined program data, and executing the program data. CP to do
U (Central Processing Unit). In this case, each program data is stored on a CD.
-It may be introduced via a storage medium such as a ROM or a flexible disk.

[0052]

According to the present invention, since the characteristic amount is calculated for each small area of the input image, the embedding strength for each area can be adjusted, and the deterioration of the image quality can be suppressed. In addition, by adjusting the embedding strength using a correspondence table corresponding to a predetermined compression bit rate, the embedding strength can be determined in consideration of the image quality of the digital watermark inserted image after image encoding.

As a result, the resistance to standard image coding such as MPEG can be improved. For example, when compression is performed at a low compression bit rate (high compression / low image quality), by increasing the embedding strength, resistance to compression can be improved. In addition, since image quality deterioration such as block distortion due to compression is more easily perceived than image quality deterioration due to embedding, even if the embedding strength is increased, image quality deterioration due to embedding can be suppressed.

On the other hand, when the compression is performed at a high compression bit rate (low compression and high image quality), deterioration of the image quality due to embedding can be suppressed by reducing the embedding strength. Moreover, since the deterioration due to compression is small, the resistance to compression can be maintained even if the embedding strength is reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram of a digital watermark embedding device according to an embodiment of the present invention.

FIG. 2 is a flowchart of the digital watermark embedding device.

FIG. 3A shows an example of the correspondence table (feature amount); (B) Illustration of the correspondence table (compression bit rate)

FIG. 4 is a block diagram of a conventional digital watermark embedding device.

[Explanation of symbols]

101 feature amount calculation unit 102 Embedding strength judgment unit 103 Digital watermark insertion unit 104 image encoding unit 401 Category Classification Unit 402 storage device 403 Digital watermark strength calculation unit 404 Digital watermark insertion unit

   ────────────────────────────────────────────────── ─── Continuation of front page    (72) Inventor Takashi Katsura             Matsushita Electric, 1006 Kadoma, Kazuma, Osaka             Sangyo Co., Ltd. F term (reference) 5B057 AA20 CA08 CA12 CA16 CB08                       CB12 CB16 CC01 CE08 DA20                       DB02 DB09 DC30 DC36                 5C053 GB07 GB21 GB37 JA30                 5C063 AB03 AC01 AC10 DA13 DA20                       DB10                 5C076 AA14 BA06

Claims (27)

[Claims] An electronic watermark insertion unit for embedding an electronic watermark in a digital image; and a processing unit for performing predetermined signal processing on the digital image in which the electronic watermark is embedded. An electronic watermark embedding device, wherein the embedding strength is determined based on the content of the signal processing of the processing unit. 2. A digital watermark embedding unit for calculating a digital image characteristic amount, wherein a digital watermark embedding strength of the digital watermark inserting unit is calculated by the signal processing content of the processing unit and the characteristic amount calculating unit. 2. The method according to claim 1, wherein the value is determined based on the characteristic amount.
Electronic watermark embedding device according to the above. 3. The feature amount calculating section calculates a feature amount for each partial area constituting a part of a digital image, and the digital watermark inserting section inserts a digital watermark according to an embedding strength determined for each partial area. 3. The digital watermark embedding device according to claim 2, wherein the electronic watermark is embedded. 4. The image processing apparatus according to claim 1, wherein the processing unit includes an image encoding unit that encodes the digital image in which the digital watermark is embedded. 4. The digital watermark embedding device according to claim 1, wherein the digital watermark embedding device is determined based on a bit rate. 5. The digital watermark embedding device according to claim 3, wherein said partial area is one or more macroblocks used in encoding of said image encoding unit. 6. A digital watermark embedding unit for embedding a digital watermark in a digital image, wherein the digital watermark embedding unit embeds a digital watermark with an embedding strength determined for each partial region constituting a part of the digital image. Watermark embedding device. 7. A feature value calculation unit for calculating a feature value of a digital image, an embedding strength determination for obtaining an embedding strength using the feature value calculated by the feature value calculation unit and a compression bit rate used in encoding. A digital watermark insertion unit that embeds a digital watermark in the digital image using the embedment strength determined by the embedment strength determination unit; and encodes the digital image in which the digital watermark insertion unit embeds a digital watermark at the compression bit rate. And an image encoding unit that outputs a compressed bit stream. 8. The digital watermark embedding device according to claim 7, wherein said characteristic amount calculating section calculates a characteristic amount for each partial region constituting a part of said digital image. 9. The electronic watermark embedding device according to claim 8, wherein the feature amount includes an average value of the luminance component in the partial region and a sum of absolute differences of horizontal and vertical adjacent pixels in the partial region. 10. The digital watermark embedding device according to claim 8, wherein the partial area is one or more macroblocks used in encoding of the image encoding unit. 11. An embedding strength determination unit, comprising: one or both of a feature amount calculated by the feature amount calculation unit and a compression bit rate used in encoding of the image encoding unit; 11. The electronic watermark embedding device according to claim 7, wherein a correspondence table is used. 12. The digital watermark according to claim 11, wherein the correspondence table is set such that the digital watermark is weakly embedded when the compression bit rate is high, and the digital watermark is strongly embedded when the compression bit rate is low. Implant device. 13. The digital watermark embedding device according to claim 7, wherein said digital watermark insertion unit embeds a digital watermark with a generally different embedding strength for each partial region constituting a part of said digital image. 14. A method for embedding a digital watermark in a digital image and performing predetermined signal processing on the digital image in which the digital watermark is embedded, wherein the embedding strength of the digital watermark is determined by the content of the predetermined signal processing. A digital watermark embedding method determined based on this. 15. A feature amount of a digital image is obtained, and embedding strength of a digital watermark is determined based on the content of the predetermined signal processing and the obtained feature amount.
Digital watermark embedding method described. 16. The digital watermark embedding method according to claim 15, wherein a characteristic amount is obtained for each partial region constituting a part of the digital image, and a digital watermark is embedded with an embedding strength determined for each partial region. 17. The method according to claim 14, wherein the predetermined process is a process of encoding a digital image in which a digital watermark is embedded, and the embedding strength of the digital watermark is determined based on a compression bit rate of the encoding. 15. The method for embedding a digital watermark according to items 16 to 16. 18. The digital watermark embedding method according to claim 16, wherein said partial area is one or more macroblocks used in said encoding. 19. Embedding a digital watermark in a digital image;
A method for embedding a digital watermark with an embedding strength determined for each partial region constituting a part of a digital image. 20. A method for calculating a feature amount of a digital image, obtaining an embedding strength using the calculated feature amount and a compression bit rate used in encoding, and using the obtained embedding strength to digitally watermark the digital image. A digital watermark embedding method, wherein a digital image in which a digital watermark is embedded is encoded at the compression bit rate, and a compressed bit stream is output. 21. The digital watermark embedding method according to claim 20, wherein a feature amount is calculated for each partial region constituting a part of the digital image. 22. The digital watermark embedding method according to claim 21, wherein the feature quantity includes an average value of luminance components in the partial area and a sum of absolute differences of horizontal and vertical adjacent pixels in the partial area. 23. A digital watermark embedding method according to claim 21, wherein said partial area is one or more macroblocks used in said encoding. 24. A digital watermark embedding method according to claim 20, wherein a correspondence table between one or both of the calculated feature amount and the compression bit rate used in the encoding and the digital watermark embedding strength is used. . 25. The digital watermark according to claim 24, wherein the correspondence table is set such that when the compression bit rate is high, the digital watermark is weakly embedded, and when the compression bit rate is low, the digital watermark is strongly embedded. Embedding method. 26. The digital watermark embedding method according to claim 20, wherein an electronic watermark is embedded with generally different embedding strengths for each partial region constituting a part of the digital image. 27. A method for calculating a feature amount of a digital image, obtaining an embedding strength using the calculated feature amount and a compression bit rate used in encoding, and electronically watermarking the digital image using the obtained embedding strength. A computer-readable recording medium storing a digital watermark embedding program for encoding a digital image with a digital watermark embedded therein at the compression bit rate and outputting a compressed bit stream.