JP4124366B2 - Method for embedding and extracting digital watermark - Google Patents

Description

  The present invention relates to the field of audio / video recording / playback apparatus using digital recording media such as CD / DVD / semiconductor memory, audio / video production, storage of materials, relay of location material, etc. The field of fine sound and video production, the field of CG animation video using computer graphics, the field of visualization video in science and technology simulation, the storage and transmission of medical sound, still images and moving images in electronic medical records and telemedicine, the genome Data such as storage and transmission of biological information such as information and protein information, storage and transmission of materials and page makeup image data in digital printing plate making processes, storage and transmission of data in the fields of remote sensing, satellite image analysis, and map information processing The present invention relates to a data processing technique suitable for a field where modification of the data is disliked.

Conventionally, a lossless compression method has been used as a method for compression-coding audio data and still image / moving image data without data loss. When encoded and distributed using the lossless compression method, high-quality audio and video can be obtained when decoded, so it is very useful when used by legitimate users. Even the person who copied it can use it. If the data encoded by the lossless compression method is diverted after decoding, the source of the data can no longer be known. Therefore, in order to prevent / suppress such unauthorized copying, there are a method of setting an encryption key at the time of decryption, and a method of inserting a digital watermark that makes it possible to trace the unauthorized use by specifying the copyright holder when decrypted. is there. There are invisible and visible types of digital watermark insertion, but the invisible type is used so as not to hinder when displaying an image (for example, see Patent Documents 1 to 4).
Japanese Patent No. 3522056 Japanese Patent Laid-Open No. 11-296661 JP 2003-174550 A JP 2004-173237 A

  However, since the above-mentioned conventional technology embeds a digital watermark directly in the original data, when trying to improve the detection performance of the invisible digital watermark at the time of decoding, the quality of the original data must be sacrificed, and the quality is affected. There is a problem that an invisible digital watermark cannot be embedded without giving a mark.

  Therefore, an object of the present invention is to provide a digital watermark embedding method and extraction method that can embed an invisible digital watermark without affecting the quality of original data.

  In order to solve the above problems, in the present invention, the encoding efficiency for each block is calculated for a block division stage for dividing the data to be encoded into a plurality of blocks and a plurality of encoding schemes prepared in advance. An encoding efficiency calculation step, a numerical array indicating invisible attribute information embedded as a digital watermark, and one encoding scheme out of the plurality of encoding schemes based on the encoding efficiency rank calculated for each block Based on the determined encoding scheme determination stage and the determined encoding scheme, encoding is performed on the block so that it is clear which encoding scheme is applied, and an encoded block is generated. The block encoding step is performed.

  According to the present invention, a plurality of encoding methods are prepared in advance, the original data is divided into a plurality of blocks, and the encoding efficiency of the prepared encoding method is calculated and ranked for each block. Each block is encoded using an encoding method that has a rank corresponding to the bit value of the invisible attribute information to be embedded as a watermark, so it is possible to embed an invisible digital watermark without affecting the quality of the original data It becomes.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
(1. Basic concept of encoding / decoding according to the present invention)
First, the basic concept of the digital watermark embedding method and extraction method in the present invention will be described. In the present invention, an invisible digital watermark is embedded at the time of encoding, and an invisible digital watermark is extracted at the time of decoding. FIG. 1 is a diagram showing the basic concept of the present invention. As shown in FIG. 1A, in encoding, original data to be encoded is divided into a plurality of blocks. Then, the encoding efficiency of a plurality of encoding methods prepared in advance for each block is calculated, and the encoding efficiency for each block is ranked. On the other hand, invisible attribute information to be embedded separately as a digital watermark is prepared, one of the encoding methods is determined based on this value and the order of encoding efficiency, and the block is determined by the determined encoding method. Is encoded. In this way, each block is encoded.

  In addition, as shown in FIG. 1B, in the decoding, each encoded block constituting the encoded data is extracted. In extracting each encoded block, the recording position at the beginning of each encoded block and the size (data amount) of each encoded block are recorded in an appropriate position in the encoded data file and read in advance. It ’s good if you keep it. Subsequently, each extracted encoded block is decoded using a decoding method corresponding to the encoding method used for encoding. In this way, each encoded block is decoded to obtain a block constituting the original original data. Then, the encoding efficiency of a plurality of encoding methods prepared in advance for each decoded block is calculated, and the encoding efficiency for each block is ranked. On the other hand, the coding scheme recorded in each coding block is extracted, this coding scheme is compared with other coding schemes, the rank is recognized, and a set of bits corresponding to the recognized rank is obtained. Obtained as invisible attribute information. The encoding as shown in FIG. 1A and the decoding as shown in FIG. 1B are performed by a computer equipped with dedicated software by an encoding device (digital watermark embedding device) and decoding device (digital watermark extraction). This is realized by functioning as a device.

(2. When the target data is acoustic data)
The present invention can be applied to various data regardless of the content and format of the data, but is mainly used for acoustic data and image data. Hereinafter, processing for these two types of data will be described. First, acoustic data will be described.

(2.1. Encoding)
In the present embodiment, analog sound signals obtained by digitizing analog sound signals by a known method such as PCM are used as the sound data. For example, when sampling is performed at a sampling frequency of 48 KHz and a quantization bit number of 16 bits (with plus and minus signs), a sample string composed of 48000 samples per second is formed, and each sample takes a value of −32768 to 32767. become.

  The encoding apparatus that implements the digital watermark embedding method according to the present invention, after reading the acoustic data as described above, blocks in units of a predetermined number of samples from the head of the sample sequence constituting the acoustic data. Read to work memory. The number of samples for one block can be set in advance. The number of samples for one block is preferably about 1 second. Even with the same length in time, the number of samples differs depending on the sampling frequency, so the number of samples that the setter is about 1 is set in the apparatus in advance. For example, when the sampling frequency of the acoustic data to be encoded is 48 kHz, 48000 samples may be set as one block in order to set one block to one second. The encoding device sequentially reads the set number of samples as one block into the work memory.

  On the other hand, as an encoding method for encoding the divided blocks, in the present embodiment, eleven linear prediction formulas [Formula 1] to [Formula 11] below are prepared.

[Formula 1]
e0 (t) = x (t) -e0 (t-1) / 2

[Formula 2]
e1 (t) = x (t ) -a 11 · x (t-1) -e1 (t-1) / 2

[Formula 3]
e2 (t) = x (t ) -a 21 · x (t-1) -a 22 · x (t-2) -e2 (t-1) / 2

[Formula 4]
e3 (t) = x (t ) -a 31 · x (t-1) -a 32 · x (t-2) -a 33 · x (t-3) -e3 (t-1) / 2

[Formula 5]
e4 (t) = x (t ) -a 41 · x (t-1) -a 42 · x (t-2) -a 43 · x (t-3) -a 44 · x (t-4) - e4 (t-1) / 2

[Formula 6]
e5 (t) = x (t ) -a 51 · x (t-1) -a 52 · x (t-2) -a 53 · x (t-3) -a 54 · x (t-4) - a ₅₅ · x (t−5) −e5 (t−1) / 2

[Formula 7]
e6 (t) = x (t ) -b 11 · x (t-1) -e6 (t-1) / 2

[Formula 8]
e7 (t) = x (t ) -b 21 · x (t-1) -b 22 · x (t-2) -e7 (t-1) / 2

[Formula 9]
e8 (t) = x (t ) -b 31 · x (t-1) -b 32 · x (t-2) -b 33 · x (t-3) -e8 (t-1) / 2

[Formula 10]
e9 (t) = x (t ) -b 41 · x (t-1) -b 42 · x (t-2) -b 43 · x (t-3) -b 44 · x (t-4) - e9 (t-1) / 2

[Formula 11]
e10 (t) = x (t ) -b 51 · x (t-1) -b 52 · x (t-2) -b 53 · x (t-3) -b 54 · x (t-4) - _{b 55 · x (t-5} ) -e10 (t-1) / 2

  In the above [Equation 1] to [Equation 11], e0 (t) to e10 (t) are prediction errors in the sample at time t according to each prediction calculation formula, and x (t) to x (t-5) are times. It is a sample value in t-t-5.

"A ₂₁ · x (t-1) + a ₂₂ · x (t-2)" in [Formula 3], "a ₃₁ · x (t-1) + a ₃₂ · x (t-) in [Formula 4] 2) + a ₃₃ · x (t−3) ”,“ a ₄₁ · x (t−1) + a ₄₂ · x (t−2) + a ₄₃ · x (t−3) + a ₄₄ · x (t-4) "," _a51 * x (t-1) + _a52 * x (t-2) + _a53 * x (t-3) + _a54 * x (t-4) " ) + A ₅₅ · x (t−5) ”,“ b ₂₁ · x (t−1) + b ₂₂ · x (t−2) ”in the above [Equation 8],“ b ₃₁ · x in the above [Equation 9] _{(t-1) + b 32} · x (t-2) + b 33 · x (t-3) ", the" b ₄₁ · x in [equation 10] _{(t-1) + b 42} · x (t-2) _{+ b 43 · x (t-} 3) + b 44 · x (t-4) ", put in [equation 11] _{"B 51 · x (t-1} ) + b 52 · x (t-2) + b 53 · x (t-3) + b 54 · x (t-4) + b 55 · x (t-5) " is the past Linear prediction component based on 2-5 samples. Using this linear prediction component and the prediction errors “e1 (t−1) / 2” to “e10 (t−1) / 2” (error feedback component) calculated in the immediately preceding sample, a prediction error at time t e0 (t) to e10 (t) are calculated.

As a fixed value in the above coefficient _{a 11 ~a 55, a 11 =} 1, a 21 = 2, a 22 = -1, a 31 = 3, a 32 = -3, a 33 = 1, a 41 = 4 , A ₄₂ = −6, a ₄₃ = 4, a ₄₄ = −1, a ₅₁ = 5, a ₅₂ = −10, a ₅₃ = 10, a ₅₄ = −5, and a ₅₅ = 1. The coefficients b _{11 to} b ₅₅ are fixed values, for example, b ₁₁ = 0.5, b ₂₁ = 1, b ₂₂ = −0.5, b ₃₁ = 1.5, b ₃₂ = −1. _{.5, b 33 = 0.5, b} 41 = 2, b 42 = -3, b 43 = 2, b 44 = -0.5, b 51 = 2.5, b 52 = -5, b 53 = 5, b ₅₄ = −2.5, and b ₅₅ = 0.5 are set.

  The encoding apparatus applies the above mathematical formulas to all samples constituting one read block and replaces the original value of x (t) with the prediction error e (t). Then, the sum of | e (t) |, which is the absolute value of the prediction error e (t), for all the samples t in one block is calculated, and the sums for the above equations are compared and ranked. The smaller this sum is, the higher the possibility that the amount of data after encoding will be small, and the higher the encoding efficiency (compression rate).

  Next, an encoding method is determined based on invisible attribute information prepared for embedding as a digital watermark. As the invisible attribute information, various information such as a character code and an image can be adopted. Whatever format is adopted, the encoding method is determined in association with the value of each bit of the bit string constituting the invisible attribute information. For example, when the value of the bits constituting the invisible attribute information is “0”, the one with the first encoding efficiency rank is selected, and when the bit value is “1”, the one with the second encoding efficiency rank is selected. The rule of selecting is set in advance. Then, the bit string of the invisible attribute information is associated with the block to be encoded, and each block is encoded by the encoding method determined according to the corresponding bit of the invisible attribute information. Here, of the above [Equation 1] to [Equation 11], one of the top two equations with high encoding efficiency is determined according to the bits of the invisible attribute information, and the processing based on the [Equation] Encoding is performed. In this embodiment, the bit string constituting the invisible attribute information is made to correspond to the encoding method in units of 1 bit, but may be made to correspond in units of 2 bits or more. For example, if the unit is 2 bits, the encoding efficiency can be up to 4th, and if it is 3 bits, the encoding efficiency can be up to 8th. At the time of encoding, information for specifying the encoding method in which the block is actually encoded is recorded in association with the encoded block or included in the block. As information for specifying the encoding method, for example, an ID may be set in advance in each of the above [formulas], and this ID may be recorded.

(2.1.1. Variable length coding)
Further, after this, a variable length encoding process for converting each sample into a variable length is performed. The variable-length coding in this embodiment employs a well-known method generally called Golomb coding. Specifically, the bit component constituting one sample is divided into an upper bit component and a lower bit component, the lower bit component is left unchanged, and the upper bit component is a numerical value obtained by decimal conversion of only the upper bit. It is an array in which bit “0” is arranged and separator bit “1” is added at the end. For example, consider an 8-bit bit component “00101000”. At this time, if the lower bit component is 4 bits, the lower bit component is “1000”. Since the upper bits are “0010”, “2” pieces of “0” obtained by decimal conversion are arranged and converted to “001” by adding “1” at the end. As a result, the 8-bit bit string “00101000” is converted into a 7-bit bit string “0011000”. In this embodiment, the bit length of the lower-order bit component that makes the bit component unchanged before and after conversion is made variable for each sample.

  In this embodiment, variable-length coding is excluded from the target of coding efficiency calculation, but in order to calculate coding efficiency more accurately, coding efficiency is calculated including variable-length coding. To do. That is, the data amounts in the state where variable length encoding is performed are compared and ranked, and it is determined that the smaller the data amount, the higher the encoding efficiency (compression rate). In this case, the accuracy is naturally high, but the entire processing time becomes long because variable length coding must be performed on each block for all of the coding schemes prepared in advance. Therefore, it is preferable to determine whether or not to perform variable-length coding for calculating the coding efficiency according to the use and urgency of the data to be coded.

  The data of each block subjected to variable length encoding as described above is sequentially recorded in one encoded file. The encoded data filed is distributed by various methods such as via a recording medium or via a communication line.

(2.2. Decoding)
In the decoding apparatus, processing means corresponding to the encoding method prepared in the encoding apparatus is prepared. For example, in the case of the above example, the same ones as [Formula 1] to [Formula 11] prepared as encoding methods are prepared. In the decoding apparatus, after reading the encoded block, information for specifying the encoding method recorded in the encoded block is read. Here, an ID for specifying [Formula] which is an encoding method is read. Subsequently, the value of each sample recorded with the prediction error in the coding block is restored to the original sample value by using [Formula] corresponding to this ID. This is executed based on an expression obtained by exchanging the term on the left side and the first term on the right side of the above [Formula] used for encoding. In this way, the original block is obtained by decoding each encoded block using the same [Formula] as the [Formula] used for encoding.

  At this point, decoding of the encoded data itself is completed, and the original data before encoding is obtained. From here, processing for detecting invisible attribute information embedded as a digital watermark is performed. Specifically, for each decoded block, first, the encoding efficiency of each of the equations [Formula 1] to [Formula 11] is calculated and ranked. The encoding efficiency calculation process is the same as the above-described encoding process. Subsequently, with reference to the ID recorded in the decoded block, for each block, the [formula] used for encoding the block is specified. Furthermore, it is recognized by referring to the already calculated encoding efficiency that the rank of [Formula] used for encoding is. Further, the value of the bits constituting the invisible attribute information is obtained from the relationship between the coding efficiency order and the bits. By executing this processing for each block, a bit string is obtained, which becomes invisible attribute information.

(3. When the target data is image data)
Next, the case where the data to be encoded is image data will be described. In the present embodiment, the image data is described assuming that the number of pixels is 640 × 480 pixels. Note that the image data can be applied to a still image itself as image data, or to image data as each frame constituting a moving image.

(3.1. Encoding)
The encoding apparatus that implements the digital watermark embedding method according to the present invention, after reading the image data as described above, blocks the pixels constituting the image data in units of a predetermined number of pixels, and stores each block in the work memory. Read. The number of pixels for one block can be set in advance. In this embodiment, 64 × 48 pixels are defined as one block, and one image data is divided into 100 blocks for processing. The encoding device sequentially reads the set number of pixels as one block into the work memory.

  On the other hand, as an encoding method for encoding the divided blocks, in the present embodiment, the following seven formulas [Formula 12] to [Formula 18] are prepared. In the present embodiment, in order to calculate the prediction error value e (i, j) of the target pixel (i, j) to be calculated, the original pixel value x (i, j) of the target pixel and FIG. The values x (i−1, j−1) and x (i) of the neighboring three pixels (i−1, j−1), (i, j−1), (i−1, j) having the positional relationship shown i, j-1) and x (i-1, j) are used.

[Formula 12]
e20 = x (i, j) -x (i-1, j)

[Formula 13]
e21 = x (i, j) -x (i, j-1)

[Formula 14]
e22 = x (i, j) -x (i-1, j-1)

[Formula 15]
e23 = x (i, j) -x (i-1, j) -x (i, j-1) + x (i-1, j-1)

[Formula 16]
e24 = x (i, j) -x (i-1, j)-{x (i, j-1) -x (i-1, j-1)} / 2

[Formula 17]
e25 = x (i, j) -x (i, j-1)-{x (i-1, j) -x (i-1, j-1)} / 2

[Formula 18]
e26 = x (i, j)-{x (i-1, j) + x (i, j-1)} / 2

  In the above [Expression 12] to [Expression 18], e20 to e26 are prediction errors of the pixel (i, j) by the respective prediction calculation expressions, and x (i, j), x (i, j-1), x (I-1, j) and x (i-1, j-1) are pixels (i, j), (i, j-1), (i-1, j), (i-1, j-1). ).

  The encoding apparatus applies the above equations to all the pixels constituting one read block and replaces the original value of x (i, j) with the prediction error e (i, j). Become. Then, the sum of | e (i, j) | that is the absolute value of the prediction error e (i, j) for all the pixels (i, j) in one block is calculated, and the sum for each of the above formulas Are compared and ranked. The smaller this sum is, the higher the possibility that the amount of data after encoding is small, and the higher the encoding efficiency (compression rate).

  Next, an encoding method is determined based on invisible attribute information prepared for embedding as a digital watermark. As in the case of the acoustic data, this is performed by determining the encoding method in association with the value of each bit of the bit string constituting the invisible attribute information.

(3.1.1. Variable length coding)
Further, after this, a variable length encoding process for converting each pixel into a variable length is performed. As in the case of the acoustic data, the variable length encoding process also employs a well-known method generally called Golomb encoding, and divides the bit component constituting one pixel into an upper bit component and a lower bit component, and then performs processing. Is done.

  Further, as in the case of the acoustic data, the encoding efficiency calculation target may be set to include variable-length encoding.

  The data of each block subjected to variable length encoding as described above is sequentially recorded in one encoded file. The encoded data filed is distributed by various methods such as via a recording medium or via a communication line.

(3.2. Decoding)
In the decoding apparatus, processing means corresponding to the encoding method prepared in the encoding apparatus is prepared. For example, in the case of the above example, the same ones as [Formula 12] to [Formula 18] prepared as the encoding method are prepared. In the decoding apparatus, after reading the encoded block, information for specifying the encoding method recorded in the encoded block is read. Here, an ID for specifying [Formula] which is an encoding method is read. Subsequently, the value of each pixel recorded with the prediction error in the coding block is restored to the original pixel value by using [Formula] corresponding to this ID. This is executed based on an expression obtained by exchanging the term on the left side and the first term on the right side of the above [Formula] used for encoding. In this way, the original block is obtained by decoding each encoded block using the same [Formula] as the [Formula] used for encoding.

  At this point, decoding of the encoded data itself is completed, and the original data before encoding is obtained. From here, processing for detecting invisible attribute information embedded as a digital watermark is performed. Specifically, as in the case of the acoustic data, first, for each decoded block, the encoding efficiencies of the formulas [Formula 12] to [Formula 18] are calculated and ranked. Subsequently, with reference to the ID recorded in the decoded block, for each block, the [formula] used for encoding the block is specified. Furthermore, it is recognized by referring to the already calculated encoding efficiency that the rank of [Formula] used for encoding is. Further, the value of the bits constituting the invisible attribute information is obtained from the relationship between the encoding efficiency rank and the bits. By executing this process for each block, a bit string is obtained, which becomes invisible attribute information.

(4.1. Embedding visible digital watermark)
As described above, in the present invention, an electronic watermark is embedded by determining an encoding method based on invisible attribute information, but in addition to this, visible attribute information can also be embedded. Next, a case where visible attribute information is added and embedded will be described. FIG. 3 is a configuration diagram of an apparatus for embedding visible attribute information. In FIG. 3, 1 is an original image input unit, 2 is a character font storage unit, 3 is an attribute information character code input unit, 4 is a raster conversion unit, 5 is a visible information embedding process execution unit, and 6 is an image output unit with a visible watermark. It is.

  The original image input unit 1 has a function of inputting original image data to be distributed. The character font storage unit 2 stores character fonts in a raster data format. The attribute information character code input unit 3 has a function of inputting attribute information to be embedded as a digital watermark in the form of a character code. The attribute information is information that allows a viewer to recognize some meaning, and includes, for example, copyright information indicating the origin of the original image and the copyright source of the original image. In addition, the digital watermark is, for example, a digital representation of a watermark that can be seen by holding a banknote over light. means. The raster conversion unit 4 has a function of converting the character code input from the attribute information character code input unit 3 into a binary data in a raster data format with reference to the character font storage unit 2. The visible information embedding processing execution unit 5 adds the original image input from the original image input unit 1 and the binary image output from the raster conversion unit 4 for each pixel, and embeds a visible digital watermark in the original image. A function of outputting as a watermarked image. The visible watermarked image output unit 7 is for outputting a watermarked image in which visible attribute information is embedded.

  Here, the process by the visible information embedding process execution unit 5 will be described with reference to FIG. 4A is an original image input from the original image input unit 1, and FIG. 4B is a binary image (identification image) that serves as a digital watermark converted by the raster conversion unit 4. . In the example of FIG. 4A, each pixel is represented by 8 bits, and each pixel takes a value of 0-255. Here, a case of a monochrome image having only one color plane will be described. FIG. 4B shows a binary image that takes only two values, 0 and 128. Here, for simplification of explanation, both the original image and the binary image are 9 × 8 pixels. When the original image shown in FIG. 4A and the binary image shown in FIG. 4B are input, the embedding processing execution unit 5 performs addition processing for each pixel. At this time, if the result of addition exceeds 255, 256 is subtracted to perform processing for accommodating 0 to 255. That is, it is expressed by 8 bits, and overflow is ignored. As a result, a watermarked image as shown in FIG. 4C is obtained. When the original image is a color image, the above addition processing with the binary image is performed for each plane of the original image.

  The apparatus shown in FIG. 3A is realized by being incorporated in one computer together with the apparatus for executing the basic concept shown in FIG. Therefore, the visible watermarked image output from the visible watermarked image output unit 7 as described above is encoded as original data in accordance with the basic concept as shown in FIG. The character code input from the attribute information character code input unit 3 is used as invisible attribute information to determine the encoding method. That is, the attribute information character code input unit 3 is used in two places as visible attribute information and invisible attribute information by inputting the attribute information only once.

(4.2. Removal of visible digital watermark)
The encoded data in which the visible attribute information and the invisible attribute information are embedded is decoded according to the basic concept shown in FIG. 1B, and the original data and the invisible attribute information are obtained. Furthermore, the visible digital watermark is removed from the visible watermarked image using the visible watermarked image and the invisible attribute information which are the original data. An apparatus for removing the visible watermark is shown in FIG. In FIG. 3B, 11 is an image input unit with a visible watermark, 12 is a character font storage unit, 13 is an attribute information character code input unit, 14 is a raster conversion unit, 15 is a watermark removal execution unit, and 16 is an original image output. Part.

  The visible watermarked image input unit 11 is for inputting a visible watermarked image that is the decoded original data. The character font storage unit 12 stores the same character font as the character font used in the visible information embedding device. The attribute information character code input unit 13 has a function of inputting invisible attribute information obtained by decoding. Here, since it is obtained in the form of a character code by decoding, it is input as it is. The raster conversion unit 14 has a function of converting the character code input from the attribute information character code input unit 13 into a binary image in a raster data format with reference to the character font storage unit 12. The watermark removal execution unit 15 adds the binary image output from the raster conversion unit 14 and the visible watermarked image input from the visible watermarked image input unit 11 for each pixel, and adds a digital watermark from the watermarked image. It has a function to output as a removed original image. The original image output unit 16 is for outputting an original image from which a digital watermark has been removed. For example, a display device that displays and outputs on a screen can be applied. The apparatus shown in FIG. 3B is implemented by being incorporated in one computer together with an apparatus that executes the basic concept shown in FIG.

  Here, the processing by the watermark removal execution unit 15 will be described with reference to FIG. As described above, the decoded image with the visible watermark is as shown in FIG. The binary image output from the raster conversion unit 14 is as shown in FIG. When the watermarked image shown in FIG. 4C and the binary image shown in FIG. 4B are input, the watermark removal execution unit 15 performs addition processing for each pixel. At this time, if the result of addition exceeds 255, 256 is subtracted to perform processing for accommodating 0 to 255. That is, it is expressed by 8 bits, and overflow is ignored. As a result, an original image as shown in FIG. When the watermarked image is a color image, the above addition processing with the binary image is performed for each plane of the watermarked image.

(5. Modifications)
The preferred embodiment of the present invention has been described above. However, the present invention is not limited to the above embodiment, and various modifications can be made. For example, in the above-described embodiment, a linear prediction formula is used as a plurality of encoding schemes. However, a known run-length scheme may be used in addition to this, or a linear prediction formula or a run-length scheme may be used. Instead of this, various known encoding methods can be used.

  In the above embodiment, the encoding efficiency is calculated after dividing the original data into a plurality of blocks. However, as preprocessing, redundant bit compression, subsample compression, and inter-channel correlation compression are performed after block division. It is also possible to combine known compression methods such as inter-frame correlation compression.

FIG. 1 is a diagram showing a basic concept of a digital watermark embedding method and extraction method according to the present invention. It is a figure which shows the positional relationship of each pixel at the time of performing a prediction error calculation process. It is a block diagram of the apparatus which performs embedding and extraction of visible information. It is a figure which shows the mode of embedding and extraction of visible information.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Original image input part 2, 12 ... Character font memory | storage part 3, 13 ... Attribute information character code input part 4, 14 ... Raster conversion part 5 ... Visible information embedding process execution part 6 ... Image output section with visible watermark 11 ... Image input section with visible watermark 15 ... Visible information removal execution section 16 ... Original image output section

Claims (9)

A block division stage for dividing the data to be encoded into a plurality of blocks;
For a plurality of coding schemes prepared in advance, an encoding efficiency calculation stage for calculating the encoding efficiency for each block;
An encoding method for determining one encoding method from among the plurality of encoding methods based on a numerical array indicating invisible attribute information to be embedded as a digital watermark and the order of encoding efficiency calculated for each block The decision stage;
Based on the determined coding method, a block coding step of performing coding on the block so as to clarify which coding method is applied and generating a coded block;
A method for embedding a digital watermark, comprising: In claim 1,
The method of embedding an electronic watermark according to claim 1, wherein the block coding step clarifies which coding method is applied by recording information specifying the coding method in the block. In claim 1 or claim 2,
Visible attribute information generating step for generating visible attribute information based on the invisible attribute information and a visible attribute for embedding the generated visible attribute information for the data to be encoded before the block division step Information embedding stage,
A method for embedding a digital watermark, comprising: In claim 3,
The data to be encoded is a recording of a still image or a moving image,
The invisible attribute information is a character code, and the visible attribute information generation step converts the character code into visible attribute information having a format of image data using a predetermined character font. Method. In claim 3,
The data to be encoded is acoustic data containing sound,
The invisible attribute information is a character code, and the visible attribute information generation step converts the character code into visible attribute information having a waveform data format using a predetermined speech synthesis method. Embed method. In claim 1 or claim 2,
The digital watermark embedding method, wherein the coding method is lossless coding using a linear prediction method, and a plurality of linear prediction equations are prepared as the respective coding methods. In claim 1 or claim 2,
The data to be encoded is a recording of a still image or a moving image,
The digital watermarking is characterized in that the block division step performs tile division so as to divide the two-dimensional space of the image data which is the still image or the image data constituting the moving image substantially equally in the X direction and the Y direction. Embedding method. A coding block recognition stage for recognizing each coding block from coding data composed of a plurality of coding blocks;
A block decoding step of decoding each encoded block based on a decoding method corresponding to the encoding method indicated to each encoded block, and restoring the block;
For a plurality of encoding schemes prepared in advance, an encoding efficiency calculation stage for calculating the encoding efficiency for each decoded block;
An invisible attribute information extraction step of extracting a numerical array indicating invisible attribute information based on the encoding efficiency rank calculated for each block and the encoding method designated by the encoding block corresponding to each block When,
A method for extracting a digital watermark, comprising: In claim 8,
Visible attribute information generated based on the invisible attribute information extracted in the invisible attribute information extraction stage is embedded in the data composed of the blocks generated in the block decoding stage,
After the invisible attribute information extraction step, for the data, a visible attribute information generation step for generating visible attribute information based on the extracted invisible attribute information,
A visible attribute information removing step of removing the generated visible attribute information from the data;
A method for extracting a digital watermark, comprising:

Images