JP4829164B2 - Image processing apparatus and image processing method - Google Patents

Description

  The present invention relates to an image processing apparatus and an image processing method for superimposing another additional sub information (security information or the like) on main image information (a person's face image or the like).

  In recent years, with the digitization of information and the spread of the Internet, techniques such as digital watermarks and digital signatures have come to be regarded as important in order to prevent forgery and alteration of images. In particular, a digital watermark technique for embedding additional sub-information in the main image information in an invisible state has been proposed as a measure against illegal copying, forgery, and falsification of a personal authentication medium such as an ID card or a photograph embedded with copyright information.

For example, there is known a digital watermark insertion method for embedding data in image data output to a printed matter by using characteristics of high spatial frequency components and color difference components that are difficult for humans to detect (see, for example, Patent Document 1). ).
Also, a digital watermark printing apparatus that can be confirmed with an optical filter is known (see, for example, Patent Document 2).

In these digital watermark embedding processing methods,
(1) Utilizing human visual characteristics
・ The gradation discrimination ability decreases as the frequency of the image increases.
-Color difference information is more difficult to distinguish than luminance information
(2) Complementary color relationship Example: Red + cyan color = achromatic color (white) (in case of additive color mixture)
(3) Applying complementary color relationship and color difference information to high frequency carrier pattern images (color difference modulation processing)
By using, it is possible to embed sub information (sub image information) in the main image information in an invisible state without causing image quality degradation.

  In the example of (2) above, red and cyan (= green + blue) are complementary colors in the case of additive color mixing, and even if red and cyan are adjacent to each other, they are discriminated by the human eye. Difficult to see and looks achromatic.

  As shown in the above example (3), the high frequency carrier pattern image is composed of a color difference grid pattern 1 in which red rich pixels 2 and cyan rich pixels 3 are repeatedly arranged as shown in FIG. The eye uses the human visual characteristic that cannot distinguish these small differences in color difference and determines that the color difference amount is plus or minus “0”.

If it is necessary to determine the authenticity of a recorded material on which an image to which the above-mentioned digital watermark technology is applied is recorded, the restoration process described in the above-mentioned patent document is performed using the key information, and the invisible state The authenticity determination is performed by restoring the sub-information recorded in (1).
More specifically, first, using a device such as a scanner or a camera, composite image information in which sub information is embedded as color difference information and recorded on a recorded material such as a card is read as digital information. Next, by applying a digital filter having a frequency component corresponding to the specific frequency information of the key information to the composite image information, the sub-information embedded as the color difference information is restored. Finally, authenticity determination is performed from the result obtained by restoration.
Japanese Patent Laid-Open No. 9-248935 JP 2001-268346 A

  However, the above-mentioned patent document discloses a method of superimposing one type of sub information on main image information. In this method, the size of the sub information is defined by the size of the main image information, and it is difficult to increase the size of the sub information. Further, since only one type of sub information is superimposed, if a recorded matter is investigated, it is noticed that the sub information is contained, leading to forgery or alteration.

  In order to solve such a problem, after embedding the first sub-information in the image information of R (red), G (green), and B (blue), cyan, magenta, yellow, There has been proposed a method of converting the four colors of black and embedding the second sub information using the image information of the four colors (see Japanese Patent Laid-Open No. 2005-159438).

  However, this method requires black ink and cannot be used in a printer without black ink. In addition, when restoring the second sub information from a recorded material having an image in which two sub information are embedded, it is necessary to irradiate infrared light that is not mounted on a general scanner or camera. The device becomes a big one.

Therefore, the present invention provides an image processing apparatus and an image that can improve the forgery / alteration resistance of a recorded matter obtained by superimposing two sub-information on main image information using two or more color difference grid patterns. An object is to provide a processing method .

An image processing apparatus according to the present invention includes an image input unit that inputs main image information including a color gradation image, and a first modulation unit that modulates the first color-difference lattice pattern using first sub-information including a binary image. A first superimposing unit that superimposes the first color-difference grid pattern modulated by the first modulating unit on the main image information input by the image input unit, and a second sub-unit composed of a binary image. According to the information, the second color difference grid pattern configured by the color information of two colors having different complementary colors from each other and having a different number of pixels from the first color difference grid pattern is modulated. synthesized by superimposing the second modulation means, the second color difference grid pattern modulated by the first and the second modulating means to the main image information obtained by superimposing the color difference grid pattern by the first superimposing means A second superposing means for creating image information, and a recording means for recording the composite image information created by the second superimposing means.

The image processing method of the present invention also includes an image input step for inputting main image information composed of a color gradation image, and a first color difference grid pattern modulated by first sub-information composed of a binary image. A modulation step, a first superimposition step of superimposing the first color-difference grating pattern modulated in the first modulation step on the main image information input in the image input step, and a second composed of a binary image. The second chrominance grid pattern composed of two colors of color information having a different complementary color relationship from the first chrominance grid pattern and having a different number of pixels from the first chrominance grid pattern. Superimposing the second color difference grating pattern modulated in the second modulation step on the main image information in which the first color difference grating pattern is superimposed in the second modulation step to be modulated and the first superposition step. By A second superimposing step of creating a formed image information, and a recording step of recording the composite image information created in the recording medium by the second superimposition steps.

According to the present invention, an image processing apparatus and an image capable of improving the forgery / alteration resistance of a recorded matter obtained by superimposing two sub-information on main image information using two or more color difference grid patterns. A processing method can be provided.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 schematically shows a configuration of an image processing system to which an image processing apparatus and an image processing method according to the present invention are applied. This image processing system is roughly divided into an image input device 11 as an image input means, a printer 12 as a recording means, a computer 13 for performing various processes, and the image input device 11, the printer 12 and the computer 13 can communicate with each other. It is comprised by the bidirectional | two-way communication means 14 and 15 connected to.

  The image input device 11 inputs a multi-tone image such as a face image on a photograph (printed material) P1 as a digital image of R (red), G (green), and B (blue). It consists of a video camera and a flatbed scanner. In this embodiment, for example, a scanner is used.

  The printer 12 outputs a recorded material P2 used as a personal authentication medium such as an ID card by recording a binary image such as a multi-tone image or a character on a recording medium. As a printer.

The computer 13 is an image processing unit that performs sub-information embedding processing for creating composite image information by embedding the sub-information in an invisible state with the naked eye of the image information input by the image input device 11. A processing unit 16 is included.
Here, the computer 13 may be, for example, a personal computer (hereinafter simply referred to as a personal computer) or a dedicated board composed of a CPU, a semiconductor memory, a hard disk device, an image capture board, and the like. In this embodiment, a personal computer is often used.
The image processing unit 16 in the present embodiment is configured by a part of an application program that operates on a personal computer.

  The bidirectional communication means 14 and 15 may be any means capable of bidirectionally communicating signals such as USB and SCSI. In this embodiment, USB 2.0 is used.

  The overall processing operation of the image processing system configured as described above will be described below with reference to the flowchart shown in FIG.

  First, a multi-tone image such as a face image is input by the image input device 11 (step S1). Here, the multi-tone image is, for example, a color face image on the photograph (printed material) P1, but is not limited thereto. The input image information is input to the image processing unit 16 of the computer 13.

  The image processing unit 16 performs a first sub information embedding process for embedding the first sub information in the input image information (step S2). Next, the image processing unit 16 performs a second sub information embedding process for embedding the second sub information in the image information in which the first sub information is embedded (step S3). In this way, the image information in which the first sub information and the second sub information are embedded is sent to the printer 12.

  The printer 12 obtains a recorded matter P2 by recording on the recording medium the image information in which the first sub information and the second sub information are embedded (step S4). Here, the first sub-information embedding process in step S2 and the second sub-information embedding process in step S3 have the same contents regarding the embedding process except that the sub-information should be embedded.

  Next, an embedding process (the processes in steps S2 and S3) for embedding the sub information in the image information in the image processing unit 16 will be described in detail.

  FIG. 3 shows a main configuration within the image processing unit 16, which includes a smoothing processing unit 21, a phase modulation processing unit 22, a color difference modulation processing unit 23, and a superimposition processing unit 24, which are personal computers. Is part of an application program that runs on For the embedding process, for example, a digital watermark processing technique described in Japanese Patent Laid-Open Nos. 11-168616 and 2001-268346 can be applied.

The main image information (embedded image information) 25 is image information on which the sub information 26 is superimposed, and corresponds to the face image of the owner in the personal authentication medium. This has 24 bits of information per pixel (8 bits for each of R, G, and B).
The sub information (embedded image information) 26 is obtained by converting the embedded information into binary image information, and corresponds to, for example, an identification number (ID number) or the like in the personal authentication medium. This has 1 bit of information per pixel.
The color difference grid pattern 27 is a grid pattern and is used as key information when the sub information 26 is restored. This has 1 bit of information per pixel.

  First, the smoothing processing unit 21 performs the smoothing process by setting the black pixel of the sub information 26 to “1” and the white pixel to “0”. Here, an area of a plurality of pixels is cut out in the horizontal direction (x direction) of the image, and a weighted average is taken.

  Next, the phase modulation processing unit 22 performs phase modulation on the color difference grating pattern 27 based on the result of the smoothing processing in the smoothing processing unit 21. Here, a 2 × 2 (black or white alternately) pixel pattern is rotated 90 degrees to the right.

  Next, the color difference modulation processing unit 23 performs color difference modulation processing using a predetermined color difference amount ΔCd based on the phase modulation result in the phase modulation processing unit 22. In this case, three components R (red), G (green), and B (blue) are calculated separately. For example, when the color difference grid pattern 27 is the color difference grid pattern 1 in which the red rich pixel 2 and the cyan rich pixel 3 as shown in FIG. 4 are repeatedly arranged, the portion corresponding to the red rich pixel 2 is R The data is + ΔCd, G, B data is -ΔCd, and the portion corresponding to the cyan rich pixel 3 is R data -ΔCd and G, B data is + ΔCd.

  Next, the superimposition processing unit 24 performs the superimposition process from the color difference modulation processing result in the color difference modulation processing unit 23 and the main image information 25, thereby creating composite image information 28. Specifically, the R, G, and B values of the color difference modulation processing result are added to the R, G, and B values of the main image information 25, respectively.

  By repeating the embedding process described above twice, it is possible to superimpose two types of sub information on the main image information 25. The composite image information 28 on which the two types of sub-information are superimposed is sent to the printer 12 and recorded on the recording medium to obtain a recorded matter P2.

  FIG. 5 and FIG. 6 show the pixel arrangement and the first sub information of the modulated first color difference grid pattern according to the present embodiment. The portion surrounded by the thick broken line in FIG. 5 shows the first color difference grid pattern 4. However, the pixel 4 a indicated by halftone dots and the pixel 4 b indicated by diagonal lines are alternately arranged in a pattern of 5 pixels. × 2 line configuration. A second color difference grid pattern 7 to be described later has 4 pixels × 2 pixels, but is finer than the first color difference grid pattern 4.

  It is known that human visual characteristics are such that the gradation difference between pixels that can be identified decreases as the resolution decreases, and the amount of color difference needs to be a value corresponding to this characteristic. For this reason, it is necessary to make the first color difference amount having a rough resolution as small as possible. In the present embodiment, the color difference amount ΔCd is set to “12” so that the first sub information can be embedded in an invisible state.

  In addition, the pixel 4a indicated by the halftone dots and the pixel 4b indicated by the oblique lines are in a complementary color relationship. In this embodiment, the pixel 4a indicated by the halftone dots is red rich, and the pixel 4b indicated by the oblique lines is a cyan complementary color. Color rich.

  Further, from the characteristics of the frequency filter used for sub-information restoration described later, the first chrominance grid pattern 4 and the second chrominance grid pattern 7 have the number of pixels 4a indicated by halftone dots and the number of pixels 4b indicated by diagonal lines (first It is preferable that the color difference grid pattern 4 of 1 is not an integral multiple of “5”). This is because when the number is an integral multiple, signal peaks in the frequency domain may be shared. Like this embodiment, a combination of even and odd numbers such as “2” and “5” is desirable.

  The color difference grid pattern groups 5a and 5b in FIG. 5 are 90 degrees out of phase with the other color difference grid pattern groups. FIG. 6 shows an example of the first sub information 6, which is a binary image composed of black cells 6 a and white cells 6 b. The color difference grid pattern groups 5a and 5b in FIG. 5 are positioned corresponding to the black cells 6a in FIG. That is, in the present embodiment, the first sub information 6 is superimposed on the main image information by modulating the phase of the color difference grating pattern.

  7 and 8 show the pixel arrangement and second sub information of the modulated second color difference grid pattern according to the present embodiment. The portion surrounded by the thick broken line in FIG. 7 shows the second color difference grid pattern 7. However, the pixel 7 a indicated by halftone dots and the pixel 7 b indicated by diagonal lines are alternately arranged in two pixels.

  As described above, since the second color difference grid pattern 7 is finer than the first color difference grid pattern 4, the amount of color difference can be increased. In the present embodiment, the color difference amount ΔCd is set to “24” so that the second sub-information can be embedded in an invisible state.

  Further, the pixel 7a indicated by halftone dots and the pixel 7b indicated by diagonal lines have a complementary color relationship. In this embodiment, the pixel 7a indicated by halftone dots is blue rich, and the pixel 7b indicated by diagonal lines is a yellow complementary color. The color richness was set to a complementary color relationship different from that of the first color difference grid pattern 4. By doing so, there is an advantage that separation becomes easier when the two sub-information is restored from the recorded matter.

  When image information is read by a scanner or the like from a recorded material on which an image on which two types of sub information are superimposed is recorded, the first sub information whose complementary color relationship is red and cyan is strongly included in the red image information. However, the second sub information whose complementary color relationship is blue and yellow is strongly included in the blue image information. For this reason, red image information can be mainly used when the first sub information is restored, and blue image information can be mainly used when the second sub information is restored.

  The color difference grid pattern groups 8a and 8b in FIG. 7 are 90 degrees out of phase with the adjacent color difference grid pattern group. FIG. 8 shows an example of the second sub information 9, but it is a binary image composed of black cells (shown by diagonal lines) 9a and white cells 9b. The color difference grid pattern groups 8a and 8b in FIG. 7 are at positions corresponding to the black cells 9a in FIG. That is, in the present embodiment, the second sub information 9 is also superimposed on the main image information by modulating the phase of the color difference grating pattern.

FIG. 9 shows a specific example of the created recorded matter P2. The recorded material P2 is, for example, a personal authentication medium such as an ID card in which a face image 30a for personal authentication and a character image (personal management information, etc.) 30b are recorded. Since the personal authentication medium needs to be counterfeited, the first sub information (stars indicated by broken lines) 31 is superimposed on the face image 30a in an invisible state by modulating the first color difference grid pattern 4. Yes. Further, the second color difference grating pattern 7 is modulated, and second sub information (heart-shaped pattern indicated by a one-dot broken line) 32 is superimposed in an invisible state. Here, the face image 30a is a color image. In addition, the star mark that is the first sub information 31 and the heart-shaped pattern that is the second sub information 32 are binary images.
In the above example, the case where both the first sub information 31 and the second sub information 32 are superimposed in an invisible state is shown, but either one may be superimposed in a visible state.

  FIG. 10 schematically shows a signal when the face image 30a of the recorded material P2 created as described above is read by an image input device such as a scanner, and the read image information is converted into a spatial frequency domain. . The horizontal axis indicates the spatial frequency in the horizontal direction of the image, and the vertical axis indicates the spatial frequency in the vertical direction of the image. The dotted circular signals are signals of the first color difference grid pattern 4 and the second color difference grid pattern 7, respectively, and can be separated as shown in FIG.

  By reading a bandpass filter having a frequency corresponding to the signal of the first color difference grid pattern 4 and applying it to the image information, it is possible to restore only the first sub information 6. Further, the second sub-information 9 can be restored by applying a band-pass filter having a frequency corresponding to the signal of the second color difference grid pattern 7 to the read image information.

  The recorded matter P2 according to the present embodiment is genuine only when both the first sub information 6 and the second sub information 9 are restored. If only one of the sub information or both sub information is not restored, it is a fake.

  As described above, according to the above-described embodiment, when superimposing another additional sub information on the main image information, the two sub information is converted into the main image using two or more color difference grid patterns. By superimposing it on the information, it is possible to improve the falsification resistance of the obtained recorded matter.

1 is a block diagram schematically showing the configuration of an image processing system to which an image processing apparatus and an image processing method according to the present invention are applied. 2 is a flowchart for explaining an overall processing operation of the image processing system in FIG. 1. The block diagram which shows roughly the principal part structure in the image processing part in FIG. The schematic diagram which shows an example of a color difference grid pattern. The schematic diagram explaining the pixel arrangement | sequence of a 1st color difference grid pattern. The schematic diagram explaining 1st subinformation. The schematic diagram explaining the pixel arrangement | sequence of a 2nd color difference grid pattern. The schematic diagram explaining the 2nd sub information. The schematic diagram which shows the specific example of the produced recorded matter. The schematic diagram which shows the frequency signal of the image information which read the face image of the produced recorded matter.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 4 ... 1st color difference grid pattern, 6 ... 1st sub information, 7 ... 2nd color difference grid pattern, 9 ... 2nd sub information, 11 ... Image input device (image input means), 12 ... Printer (recording) Means), 13: Computer, 14, 15: Two-way communication means, 16: Image processing section (image processing means), 21: Smoothing processing section, 22: Phase modulation processing section, 23: Color difference modulation processing section, 24: Superposition Processing unit 25 ... main image information, 26 ... sub-information, 27 ... color difference grid pattern.

Claims (2)

Image input means for inputting main image information consisting of color gradation images ;
First modulation means for modulating the first color-difference grating pattern with first sub-information consisting of a binary image ;
A first superimposing unit that superimposes the first color difference grid pattern modulated by the first modulating unit on the main image information input by the image input unit;
Due to the second sub-information consisting of a binary image , the first color-difference grid pattern is different from the first color-difference grid pattern, and is composed of color information of two colors having a complementary color relationship different from each other. Second modulation means for modulating the second color difference grating pattern;
Secondly, composite image information is created by superimposing the second color difference grid pattern modulated by the second modulation means on the main image information on which the first color difference grid pattern is superimposed by the first superimposing means . Superimposing means,
Recording means for recording the composite image information created by the second superimposing means on a recording medium;
An image processing apparatus comprising: An image input process for inputting main image information composed of color gradation images;
A first modulation step of modulating the first color-difference grating pattern with first sub-information consisting of a binary image;
A first superimposing step of superimposing the first color difference grid pattern modulated by the first modulation step on the main image information input by the image input step;
Due to the second sub-information consisting of a binary image, the first color-difference grid pattern is different from the first color-difference grid pattern, and is composed of color information of two colors having a complementary color relationship different from each other. A second modulation step of modulating the second color difference grating pattern;
Secondly, composite image information is created by superimposing the second color difference lattice pattern modulated by the second modulation step on the main image information on which the first color difference lattice pattern is superimposed by the first superposition step. A superimposing process of
A recording step of recording the composite image information created by the second superimposition step on a recording medium;
An image processing method comprising:

Images