JP4822340B2 - Color image compression encoding method, decoding method, color image compression encoding device, and decoding device - Google Patents

Description

  The present invention relates to a color image compression encoding method, a decoding method, a color image compression encoding apparatus, and a decoding apparatus that compress and encode a color image such as a color still image or a color moving image.

In recent years, a color image signal captured by a solid-state imaging device such as a CCD is converted into a digital signal, and the converted image is compressed and encoded by a conversion method such as two-dimensional orthogonal transformation, and an image storage medium such as a semiconductor memory or a hard disk. An image processing apparatus such as a digital camera for recording in a digital camera has been put into practical use.
In such an image processing apparatus, when transmitting or storing a color image, a process for compressing and encoding the color image is performed in order to suppress an enormous amount of information. Conventionally, as an image compression method, an image is divided into blocks for each of a plurality of pixels, represented by a JPEG (Joint Photographic Expert Group) method, and then subjected to predetermined conversion for each block and then quantized. In general, a compression method is used.
For example, a color image composed of R (red), G (green), and B (blue) color signals is converted into a luminance signal and a color difference signal, the luminance signal is converted into a binary value, and a luminance binary signal is generated. The color difference signal is converted into a ternary value to generate a chrominance ternary signal, and the luminance binary signal and the chrominance ternary signal are encoded so that the R, G, and B color signals of the color image are encoded. Color image encoding method (see Patent Document 1) that performs encoding after removing the correlation between the luminance signal and the color difference signal, and detects the saturation of the image signal from the absolute value of the color difference signal A color image encoding device (see Patent Document 2) that controls the quantization step width in accordance with the detected saturation has been proposed.
Japanese Patent Laid-Open No. 5-328141 (for example, claims 1, (0016), (0026) to (0030), FIG. 1, etc.) JP-A-7-107515 (for example, claim 1, (0016) and FIG. 1)

  By the way, with the recent spread of mobile devices such as mobile phones, laptop computers, and digital cameras, compression ratios have been increased in order to record a large amount of image information in a small memory device such as a memory card and a relatively small storage capacity. The demand for high compression methods is increasing. The conventional color image compression methods in Patent Document 1 and Patent Document 2 can perform highly efficient compression coding. However, since the luminance signal and the color difference signal are independently encoded, the color image information There is a limit to reducing the amount of information in transmission, display, storage, etc.

  The present invention solves such problems, and further reduces the amount of information in transmission, display, storage, etc. of color image information, and stores a large amount of color images even in a small memory device having a small storage capacity. The present invention also provides a color image compression encoding method, decoding method, color image compression encoding device, and decoding device capable of transmitting and displaying color image information with a small amount of information.

Compression coding method of a color image of the present invention according to claim 1, a compression coding method of a color image compression-encoding a color image, the digital watermark in the luminance component of the color difference component of the original image according to a predetermined rule embedded by, characterized in that compression coding it.
According to a second aspect of the present invention, in the color image compression encoding method according to the first aspect, the low frequency component of the color difference component is embedded at the position of the frequency component that does not overlap the low frequency component of the color difference component in the luminance component. It is characterized by that.
According to a third aspect of the present invention, in the color image compression encoding method according to the first or second aspect, the low frequency component of the color difference component is embedded in the high frequency component of the luminance component.
According to a fourth aspect of the present invention, in the color image compression encoding method according to the first or second aspect, the low frequency component of the color difference component is embedded in the intermediate frequency component of the luminance component.
According to a fifth aspect of the present invention, in the color image compression encoding method according to any one of the first to fourth aspects, all of the color difference component blocks are embedded after the low frequency component of the color difference component is embedded in the luminance component. It is characterized in that the component of is filled with zero.
The color image decoding method of the present invention according to claim 6 is a color image decoding method for decoding a signal obtained by compressing and encoding a color image, wherein the color difference component of the original image is converted into a luminance component according to a predetermined rule. The digital watermark is embedded, the signal embedded in the digital watermark and compressed and encoded is received, and the digital watermark is canceled according to the predetermined rule.
A color image compression coding apparatus according to a seventh aspect of the present invention is a color image compression coding apparatus for compression coding a color image, wherein the first DCT performs discrete cosine transform on a luminance signal, and the first DCT. A first quantizer that quantizes the output of one DCT, a second DCT that performs discrete cosine transform on a color difference signal, a second quantizer that quantizes the output of the second DCT, A key file describing a rule for embedding the color difference signal output from the second quantizer in the luminance signal output from the first quantizer, and the color difference signal according to the rule described in the key file; An electronic watermark processing device that embeds in a luminance signal and performs digital watermark processing, and an encoder that encodes an output of the digital watermark processing device.
The color image decoding apparatus according to claim 8 is a color image decoding apparatus for decoding a signal obtained by compressing and encoding a color image, wherein the color difference component of the original image is represented by a luminance component according to a digital watermark rule. the embeds an electronic watermark, and a decoder for decoding the compressed encoded signal embedded by the electronic watermark, and the key file containing the rule of the electronic watermark, an output of said decoder to said key file A digital watermark canceling device for canceling the digital watermark according to the rules described, a first inverse quantizer for canceling the quantization of the quantized luminance signal released by the digital watermark canceling device, A first inverse DCT that performs an inverse discrete cosine transform on the output of one inverse quantizer, and a second that dequantizes the quantized chrominance signal that has been canceled by the digital watermark cancellation apparatus. And inverse quantizer, and having a second inverse DCT to the inverse discrete cosine transform output of the second inverse quantizer.
The recording medium of the present invention according to claim 9 records the algorithm in the color image compression encoding method of claim 1 or claim 6 or the color image decoding method of claim 6. To do.

According to the present invention, when a color image is compression-encoded, since the color difference component is embedded in the luminance component by digital watermarking, it is not necessary to independently encode the luminance signal and the color difference signal. Therefore, the amount of color image information can be further reduced as compared with a color image compression encoding method using a conventional image compression method such as JPEG.
In addition, a very large amount of color images can be stored even when recorded and stored in a small memory device having a small storage capacity.
Also, color image information can be transmitted and displayed with an extremely small amount of information.
In addition, by embedding the color difference component at the position of the luminance component that is difficult to visually perceive, color image compression with little image degradation can be performed.
Further, since the color difference component is embedded in the luminance component by digital watermark, the color difference component can be used as authentication information.

The color image compression encoding method according to the first embodiment of the present invention embeds a color difference component of an original image in a luminance component according to a predetermined rule with a digital watermark , and compresses and encodes this . According to the present embodiment, it is not necessary to independently encode the luminance signal and the color difference signal. Therefore, the amount of color image information can be further reduced as compared with a color image compression encoding method using a conventional image compression method such as JPEG. In addition, a very large amount of color images can be stored even when recorded and stored in a small memory device having a small storage capacity. Also, color image information can be transmitted and displayed with an extremely small amount of information. Further, since the color difference component is embedded in the luminance component by digital watermark, the color difference component can be used as authentication information.
According to the second embodiment of the present invention, in the compression coding method according to the first embodiment, the low frequency component of the color difference component is embedded in the position of the frequency component that does not overlap the low frequency component of the color difference component in the luminance component. Is. According to the present embodiment, the color difference component and the luminance component can be reliably separated.
The third embodiment of the present invention embeds the low frequency component of the color difference component in the high frequency component of the luminance component in the compression encoding method according to the first or second embodiment. According to the present embodiment, the color difference component and the luminance component can be reliably separated. In addition, since the color difference component is embedded in the high-frequency component of the luminance component that is difficult to visually perceive, color image compression with little image degradation can be performed.
The fourth embodiment of the present invention embeds the low frequency component of the color difference component in the intermediate frequency component of the luminance component in the compression encoding method according to the first or second embodiment. According to the present embodiment, the color difference component and the luminance component can be reliably separated. Further, since the color difference component is embedded in the intermediate frequency component of the luminance component that is relatively difficult to visually perceive, color image compression with little image degradation can be performed.
According to the fifth embodiment of the present invention, in the compression encoding method according to the first to fourth embodiments, after embedding the low frequency component of the color difference component in the luminance component, all the components of the color difference component block are zeroed. It will be filled with. According to the present embodiment, the color difference component block is not necessary, and the number of blocks used as JPEG is 4. Therefore, the information amount for 2 blocks is reduced.
Decoding method of a color image according to the sixth embodiment of the present invention, embeds the digital watermark in the luminance component by a color difference component of the original image a predetermined rule, the compressed encoded signal embedded by the electronic watermark The digital watermark is received according to a predetermined rule. According to the present embodiment, a color image can be reliably restored from a monochrome image in which the color difference component is hidden.
The color image compression coding apparatus according to the seventh embodiment of the present invention uses a key file that describes a rule for performing discrete cosine transform and quantization on a luminance signal and a color difference signal independently and embedding the color difference signal in the luminance signal. Then, according to the rule, the color difference signal is embedded in the luminance signal and digital watermark processing is performed for encoding. According to the present embodiment, it is possible to embed a color difference signal in a luminance component and perform digital watermark processing by slightly improving a conventional image compression apparatus such as a JPEG encoder.
Decoding device of a color image according to the eighth embodiment of the present invention, the color difference component of the original image embeds an electronic watermark in the luminance component by the electronic watermark decoding compressed encoded signal embedded by the electronic watermark Then, the digital watermark is released according to the digital watermark rule to separate the luminance signal and the color difference signal, the quantization of each of the luminance signal and the color difference signal is released, and the inverse discrete cosine transform is performed. According to the present embodiment, a signal obtained by embedding a color difference signal in a luminance component and digitally watermarking it can be decoded only by slightly improving a conventional image decoding device such as a JPEG decoder.
A recording medium according to the ninth embodiment of the present invention records an algorithm in the first to fifth color image compression encoding methods or the sixth color image decoding method. According to the present embodiment, a recording medium can be used as a part of various color image processing apparatuses without using hardware such as a color image compression encoding apparatus and decoding apparatus, and input / output of a personal computer. By using it in the section, the color image compression encoding method and decoding method of the present invention can be implemented.

A color image compression encoding method and apparatus according to an embodiment of the present invention will be described below with reference to the drawings. In the following embodiment, an embodiment in which the present invention is applied to the JPEG system will be described.
FIG. 1 is a conceptual diagram illustrating the principle of a color image compression encoding method and decoding method according to an embodiment of the present invention, and FIG. 2 is a flowchart illustrating the operation of the color image compression encoding method in FIG. .
The color signals R (red), G (green) and B (blue) signals read from the original image 1 are subjected to image color conversion, and the luminance signal Y and the color difference signals Cr (= R−Y), Cb (= B -Y) (step 1). The converted luminance signal Y and color difference signals Cr and Cb are supplied to the JPEG encoder 2 with a digital watermark function. In the JPEG encoder 2 with a digital watermark function, the luminance signal Y is subjected to discrete cosine transform (hereinafter abbreviated as DCT) (step 2) and then quantized (step 3). Similarly, the color difference signals Cr and Cb are DCTed (step 4) and then quantized (step 5). The luminance signal Y quantized in step 3 and the color difference signals Cr and Cb quantized in step 5 are supplied to the digital watermark device in the JPEG encoder 2 with digital watermark function, and are specified in advance in the key file 3. For example, the color difference signals Cr and Cb are inserted into the high frequency component of the luminance signal Y as electronic watermarks (step 6). Details of the key file 3 and the rules specified thereto will be described later. As a result, a signal in which the chrominance signal component is embedded in the luminance signal component by the digital watermark technique is obtained, and this signal is entropy encoded (step 7) to be encoded as a monochrome JPEG image 4 in which the chrominance component is hidden. Thus, from the output of the JPEG encoder 2 with the digital watermark function, a monochrome signal in which the color signal in the original image 1 is hidden is output as a monochrome JPEG image 4 compressed by the JPEG method (step 8).

Next, the key file 3 and the rules specified thereto will be described. The key file 3 is a rule that defines which frequency component of the color difference signals Cr and Cb is embedded in which frequency component of the luminance signal Y at the time of encoding by digital watermark processing, and how at the time of decoding by the digital watermark cancellation processing This file describes the rules that specify whether to restore.
In general JPEG, if the DCT block is 8 × 8 samples, as shown in FIG. 3B, the Y signal of 16 × 16 samples includes four Y component blocks Y1, Y2, Y3, and Y4, It is decomposed into one Cr component block and one Cb component block. The DCT and quantization described in FIG. 2 are performed on each of these blocks.
The number of each sample in FIG.3 (b) shows the number corresponding to a DCT coefficient. The description of the key file describes the conversion source number and the conversion destination number in the CSV format of two columns as shown in FIG. In other words, the position number corresponding to the Cr and Cb components to be embedded is described in the conversion source number column, and the position number corresponding to the Y component to be embedded is described in the conversion destination number column.

FIG. 4 shows an embodiment in which the low frequency components of the Cr and Cb components are embedded in the position of the high frequency component which is the position of the frequency component that does not overlap with the low frequency components of the Cr and Cb components in the Y component. As shown in FIG. 4A, the conversion source number and the conversion destination are such that the low frequency component of the Cb component is embedded in the high frequency component of the Y1 and Y2 blocks, and the low frequency component of the Cr component is embedded in the high frequency component of the Y3 and Y4 blocks. The number is written. By this rule, embedding is performed in the Y component of the Cr and Cb components as shown in FIG. After the low frequency components of the Cr and Cb components are embedded at the position of the high frequency component of the Y component, the Cr component of the Cr block and the Cb component of the Cb block are all filled with zeros and discarded. Therefore, since the blocks used as JPEG are four blocks, the information amount for two blocks is reduced.
Since the high-frequency component of the luminance component is difficult to be visually perceived by humans, color image compression with little image degradation can be performed according to this embodiment.
Further, since the color difference component is embedded in the luminance component by digital watermark, the color difference component can also be used as authentication information.

FIG. 5 shows an embodiment in which the low frequency components of the Cr and Cb components are embedded in the position of the intermediate frequency component that is the position of the frequency component that does not overlap with the low frequency components of the Cr and Cb components in the Y component. As shown in FIG. 5A, the low frequency component of the Cb component is embedded in the intermediate frequency component of the Y1 and Y2 blocks, and the low frequency component of the Cr component is embedded in the intermediate frequency component of the Y3 and Y4 blocks. The conversion destination number is described. By this rule, the Y component of the Cr and Cb components is embedded as shown in FIG. After the low frequency components of the Cr and Cb components are embedded at the position of the intermediate frequency component of the Y component, the Cr component of the Cr block and the Cb component of the Cb block are all filled with zeros and discarded. Also in this embodiment, since the blocks used as JPEG are four blocks, the information amount for two blocks is reduced.
Also in this embodiment, since the intermediate frequency component of the luminance component is relatively difficult to perceive in human vision, color image compression with relatively little image degradation can be performed.
Further, since the color difference component is embedded in the luminance component by digital watermark, the color difference component can also be used as authentication information.

Next, the color image decoding method in FIG. 1 will be described with reference to the flowchart of FIG.
FIG. 6 is a flowchart for explaining the operation of the color image decoding method in FIG. 1. FIG. 6A is a flowchart in the case of decoding the monochrome JPEG image 4 obtained in step 8 of FIG. (B) is a flowchart in the case of decoding the monochrome JPEG image 4 with the JPEG decoder 7 with a digital watermark cancellation | release function.
In FIG. 6A, a monochrome JPEG image 4 is input to a commonly used JPEG decoder 5 (step 11) and decoded by the JPEG decoder 5 (step 12). The decoded signal is inversely quantized (step 13), further subjected to inverse discrete cosine transform (step 14), and the restored monochrome image 6 is restored (step 15). That is, the signal decoded in step 12 is not subjected to digital watermark cancellation of the color difference signals Cr and Cb embedded with the digital watermark, so that the color difference signal component is not restored and only the luminance signal Y is restored.
As described above, in the digital watermark, when the Cr and Cb components are embedded in the Y component, the Y component at the position where the Cr and Cb components are embedded is replaced with the Cr and Cb components, and the frequency component at the position is not reproduced. Therefore, the restored monochrome image 6 is a gray scale image in which noise is included in the luminance component.

On the other hand, as shown in FIG. 6B, when the monochrome JPEG image 4 according to the present invention is input to the JPEG decoder 7 with a digital watermark cancellation function (step 21), the monochrome JPEG image 4 is converted into the JPEG decoder 7 with a digital watermark cancellation function. (Step 22). At this time, along with the decoding of the luminance signal Y, the digital watermark cancellation of the color difference signals Cr and Cb embedded in the digital watermark is performed according to the digital watermark cancellation rule specified in the key file 3. Therefore, the signal decoded in step 22 is output as a color signal. This color signal is inversely quantized (step 23), and further subjected to inverse discrete cosine transform (step 24) to restore the restored color image 8 (step 25). As described above, when decoding is performed by the JPEG decoder 7 with the digital watermark cancellation function, the digital watermark of the color signal component is restored, and the color difference signals Cr and Cb are restored simultaneously with the luminance signal Y, and the restored color image 8 is output. .
As described above, in the digital watermark, when the Cr and Cb components are embedded in the Y component, the Cr component of the Cr block and the Cb component of the Cb block are all padded with zeros and discarded, and the Cr and Cb components are discarded. Since the Y component at the position where is embedded is replaced by the Cr and Cb components and the frequency component at that position is not reproduced, the restored color image 8 is partially missing the frequency component compared to the original image 1. However, there is no practical problem if the missing frequency component is set to a component that has little visual impact.

FIG. 7 is a block diagram of an embodiment of the color image compression encoding apparatus used in the present invention, and corresponds to the JPEG encoder 2 with digital watermark function in FIG.
R (red), G (green) and B (blue) signals which are color signals read from the original image 1 are converted into luminance signal Y and color difference signals Cr (= R−Y), Cb (= B) by the color conversion device 11. -Y). The luminance signal Y is subjected to discrete cosine transformation by a discrete cosine transformer (hereinafter referred to as DCT) 12. As shown in FIG. 4, if the DCT block is 8 × 8 samples, the 16 × 16 sample Y signal is divided into four blocks Y1, Y2, Y3, and Y4. On the other hand, the Cr signal and the Cb signal are respectively 2 blocks of 8 × 8 samples of Cr and Cb, and one block of Cr and one block of Cb with respect to four blocks of Y1, Y2, Y3, and Y4 in the Y signal. Corresponds. Signal processing such as quantization is performed in units of this combination. The coefficients subjected to the discrete cosine transform by the DCT 12 are supplied to the quantizer 13, quantized with a step width given from a quantization controller (not shown), and supplied to the digital watermark processing device 14.
On the other hand, the Cr signal and the Cb signal are subjected to discrete cosine transform by the DCT 15 and quantized by the quantizer 16. The quantized Cr signal and Cb signal are supplied to the digital watermark processing device 14 according to the rules specified in the key file 3 with reference to the key file 3.
In the digital watermark processing apparatus 14, the Cr signal and the Cb signal are embedded at the sample position in a predetermined block of the Y signal according to the rule specified by the key file 3. As a result, the color difference signal is embedded as a digital watermark in the luminance signal, and the output from the digital watermark processing device 14 is output as a four-block Y signal including the encrypted color difference signal. The output from the digital watermark processing apparatus 14 is supplied to the encoder 18 where encoding such as entropy encoding is performed and an encoded signal with digital watermark is output. In the digital watermarked encoded signal, signals of one block of Cr and one block of Cb corresponding to the color difference signal are compressed, and output as an encoded signal of only a luminance signal at first glance.

FIG. 8 is a block diagram of an embodiment of the color image decoding apparatus used in the present invention, and corresponds to the JPEG decoder 7 with digital watermark cancellation function in FIG.
The digital watermarked encoded signal output from the encoder 18 in the color image compression encoding apparatus of FIG. 7 is supplied to the decoder 21 and decoded. The decrypted signal is supplied to the digital watermark cancellation processing device 22, and the digital watermark of the Cr signal and the Cb signal embedded in the Y signal is canceled according to the rule specified in the key file 3, and the luminance signal Y and the color difference The signals Cr and Cb are restored. The luminance signal Y from which the digital watermark has been restored is inversely quantized by the inverse quantizer 24, further subjected to inverse discrete cosine transform by the inverse DCT 25 and supplied to the inverse color converter 26. Similarly, the color difference signals Cr and Cb whose digital watermark has been restored are inversely quantized by the inverse quantizer 27, further subjected to inverse discrete cosine transform by the inverse DCT 28 and supplied to the inverse color converter 26. The reverse color converter 26 converts the input luminance signal Y and color difference signals Cr and Cb into R, G, and B primary color image signals.

  7 and 8, the example in which the color image compression encoding method and decoding method of the present invention are implemented by the color image compression encoding device and decoding device has been described. However, the color image of the present invention has been described. The compression encoding method and the decoding method algorithm can be programmed and stored in various recording media such as a semiconductor chip such as a CPU and a memory element.

As described above, according to the present invention, when a color image is compressed and encoded, the chrominance component is embedded in the luminance component with a digital watermark, so that the luminance signal and the chrominance signal need to be encoded independently. Absent. Therefore, the amount of color image information can be further reduced as compared with a color image compression encoding method using a conventional image compression method such as JPEG. In addition, a large amount of color images can be stored in a small memory device having a small storage capacity, and color image information can be transmitted and displayed with a small amount of information.
In addition, by embedding the color difference component in the high frequency and intermediate frequency components of the luminance component that are difficult to visually perceive, color image compression with little image degradation can be performed.
Further, since the color difference component is embedded in the luminance component by digital watermark, the color difference component can be used as authentication information.
In the above description, the present invention has been described by taking the JPEG method as an example, but it can also be applied to various image compression methods such as MLPEG, MPEG, and DV.

The color image compression encoding method, decoding method, color image compression encoding device, and decoding device of the present invention can be applied as various image compression methods such as JPEG, MLPEG, MPEG, and DV. . For example, a color image is obtained by compressing and encoding a color image signal imaged by a solid-state imaging device such as a CCD and recording and storing it in an image storage medium such as a semiconductor memory, a memory card or a hard disk. Records and accumulates with a small amount of information.
In addition, by applying to color image compression when sending and receiving color images using mobile devices such as mobile phones, laptop computers, and digital cameras, the amount of information in the transmission, display, and storage of color image information is greatly increased. Can be reduced.

1 is a conceptual diagram illustrating the principle of a color image compression encoding method and decoding method according to an embodiment of the present invention. Flowchart for explaining the operation of the color image compression coding method in FIG. It is a figure explaining the concept of the key file used for the Example of this invention, (a) is a description example of the rule of a digital watermark, (b) is Cr, the low frequency component of Cb component, Cr in Y component, An example in the case of embedding in the position of the high frequency component which is the position of the frequency component which does not overlap with the low frequency component of Cb component FIG. 5 is a diagram illustrating an example of a key file used in an embodiment of the present invention, where (a) is a description example of a rule for embedding a low-frequency component of a color difference component at a position of a high-frequency component of a luminance component, and (b) is a diagram Conceptual diagram of the DCT block when executing the rules described in (a) FIG. 5 is a diagram showing an example of a key file used in an embodiment of the present invention, where (a) is a rule description example when embedding a low frequency component of a color difference component at a position of an intermediate frequency component of a luminance component; Is a conceptual diagram of a DCT block when executing the rule described in (a) FIG. 2 is a flowchart for explaining the operation of the color image decoding method in FIG. 1, (a) is a flowchart in the case of decoding by a normal JPEG decoder, and (b) is a case of decoding by a JPEG decoder with a digital watermark release function according to the present invention. Flowchart The block diagram in the Example of the compression encoding apparatus of the color image used for this invention The block diagram in the Example of the decoding apparatus of the color image used for this invention

DESCRIPTION OF SYMBOLS 1 Original image 2 JPEG encoder with digital watermark function 3 Key file 4 Monochrome JPEG image 5 JPEG decoder 6 Restored monochrome image 7 JPEG decoder with digital watermark release function 8 Restored color image 11 Color conversion device 12, 15 DCT
13, 16 Quantizer 14 Digital watermark processing device 21 Decoder 22 Digital watermark cancellation processing device 24, 27 Inverse quantizer 25, 28 Inverse DCT
26 Reverse color converter

Claims (9)

A color image compression encoding method for compressing and encoding a color image, wherein a color difference component of an original image is embedded in a luminance component with a digital watermark according to a predetermined rule, and this is compressed and encoded Compression encoding method.   2. The color image compression encoding method according to claim 1, wherein the low frequency component of the color difference component is embedded at a frequency component position that does not overlap with the low frequency component of the color difference component in the luminance component.   3. The color image compression encoding method according to claim 1, wherein the low frequency component of the color difference component is embedded in the high frequency component of the luminance component.   3. The color image compression encoding method according to claim 1, wherein the low frequency component of the color difference component is embedded in the intermediate frequency component of the luminance component.   5. The color image compression encoding method according to claim 1, wherein after the low-frequency component of the color difference component is embedded in the luminance component, all the components of the color difference component block are embedded with zeros. . A decoding method of a color image of decoding the compression encoded signal a color image, embeds the digital watermark in the luminance component by a color difference component of the original image a predetermined rule, it is compressed and encoded by embedding an electronic watermark A method of decoding a color image, comprising: receiving a received signal and canceling the digital watermark according to the predetermined rule. A color image compression coding apparatus for compressing and coding a color image, a first DCT for discrete cosine transform of a luminance signal, and a first quantizer for quantizing an output of the first DCT; A second DCT for discrete cosine transform of the color difference signal, a second quantizer for quantizing the output of the second DCT, and the color difference signal output from the second quantizer for the first DCT A key file that describes a rule to be embedded in a luminance signal output from a quantizer, a digital watermark processing device that embeds the color difference signal in the luminance signal according to the rule described in the key file, and performs digital watermark processing; A color image compression encoding apparatus comprising: an encoder for encoding an output of a watermark processing apparatus. A decoding apparatus of a color image for decoding the compression encoded signal a color image, embeds the digital watermark in the luminance component by the original image watermarking rules the color difference components of the compression coding by embedding an electronic watermark A decoder that decodes the received signal, a key file that describes the rule of the digital watermark, and a digital watermark canceling device that cancels the digital watermark from the output of the decoder according to the rule described in the key file; , A first inverse quantizer that dequantizes the quantized luminance signal that has been canceled by the digital watermark cancellation apparatus, and a first that performs inverse discrete cosine transform on the output of the first inverse quantizer The inverse DCT, a second inverse quantizer that dequantizes the quantized color difference signal that has been canceled by the digital watermark canceling device, and an output of the second inverse quantizer Decoding device of a color image, characterized in that a second inverse DCT for cosine transform.   6. A recording medium on which an algorithm in the color image compression encoding method according to claim 1 or the color image decoding method according to claim 6 is recorded.