JP4232689B2 - Information embedding method and information extracting method - Google Patents

Description

  The present invention relates to an information embedding method and an information extracting method, and more particularly to an information embedding method and an information extracting method characterized by image combination by image processing.

  As an information embedding and information extracting technique using a digital watermark technique, there is a technique disclosed in Japanese Patent Laid-Open No. 2003-101762 “Watermark information embedding apparatus and watermark information extracting apparatus” (Oki Electric Industry Co., Ltd.). This is a technique capable of extracting information embedded as a digital watermark by imaging and processing a digital watermarked document printed on paper with a flatbed scanner or the like. However, in order to extract information with this technology, it is necessary to image and process all the objects to be read. For example, information embedded in the entire A4 paper has a narrow viewing angle such as a camera mounted on a mobile phone. Could not read from a small image taken with the device.

  On the other hand, there is a technique disclosed in Japanese Patent Laid-Open No. 2001-189889 “Image processing apparatus and method thereof, control apparatus and control method of imaging apparatus, and storage medium” (Canon Inc.). According to this technique, it is possible to efficiently generate a combined image by capturing a plurality of partial images and combining a plurality of average images to generate a wide area image. This technique is generally referred to as mosaicing technology. By detecting and synthesizing highly correlated parts from a small number of shot images by template matching, etc., a high-resolution and wide-range image can be obtained. Enables synthesis.

  Note that the method disclosed in Japanese Patent Application Laid-Open No. 2003-101762 can extract information even from a small image taken with a portable camera or the like by using this mosaicing technology together.

JP 2003-101762 A JP 2001-189889 A JP 2003-209676 A

However, the digital watermark reading method used in combination with the mosaicing technique has the following problems.
-Since a portion having a high correlation between images is detected and combined, the captured images at completely different positions cannot be correctly combined, and a correct image cannot be obtained.
-Whether the obtained image is sufficient in terms of information extraction will not be known without trying information extraction.
・ If information extraction fails, it is not clear whether the image can be extracted in the first place or only the shooting part is insufficient.
・ If the image is not enough to extract information due to lack of shooting parts, it is difficult to know which part of the image needs to be shot.
-When the subject is cut or cut off, the image cannot be combined or extracted if it is an incomplete watermark document.
・ When the amount of information is small, it is redundant to capture the entire image, but even in such a case, it is necessary to capture the entire image.

  The present invention has been made in view of the above-described problems of conventional information embedding and information extraction techniques, and an object of the present invention is to combine a plurality of partial images in an accurate arrangement or to combine information. To provide an information embedding method and an information extracting method capable of accurately restoring embedded information when an embedded image is divided into a plurality of partial areas.

  In order to solve the above problems, according to a first aspect of the present invention, there is provided an information embedding method for embedding information in an image using a digital watermark, the step of inputting an image (S101), and the input image And a step of embedding information for specifying a position on the image (S103), an information embedding method is provided (claim 1).

  Various information may be embedded in the image. That is, another information embedding method of the present invention is an information embedding method for embedding information in an image using a digital watermark, the step of inputting an image, the step of inputting embedding information embedded in the input image, And a step of embedding embedding information in a minute region of the input image (claim 2).

  According to such a method, position information is embedded in a small area unit of a background of a printed matter or a screen display by digital watermark technology, and a plurality of images can be accurately arranged based on position information obtained from a watermarked photographed image. Can be combined. In this way, a combined image can be obtained with an accurate arrangement even if there are images that are not connected to the shooting area or if the shot images are missing or overlapped in a plurality of shot images. Even when the object to be photographed is cut, the arrangement before cutting can be reproduced with an accurate arrangement. In addition, by dividing information into small area units and assigning identification numbers to the information in each small area unit, even if an image with embedded information is divided into multiple areas, it is divided based on the identification number. Previous information can be restored. Even if an area is missing, error correction coding is applied to the information before division to provide redundancy, and information divided into small areas is embedded multiple times in different locations, so that the embedded information is not lost. Can be restored. For example, by embedding the information of the printer at the time of printing, it is possible to track the identification of the printer even when the shooting target is partially cut and diverted.

  In the information embedding method of the present invention, the following applications are possible.

  Furthermore, a step (S104) of printing an image in which information for specifying a position on the image or embedded information is embedded may be included (claim 3).

  In the case of including the step of printing an image, the method may further include a step (S200) of dividing an image in which information for specifying a position on the image or embedded information is embedded (claim 200). In the step of dividing, the image may be divided so as to be equal to or smaller than an area that can be output by one sheet in the step of printing the image. According to this method, even if the captured image has a low resolution, a high resolution / high definition image can be formed by combining a plurality of captured images.

  Also, in the step of dividing the image, the reading area (for example, scanner, digital camera, camera-equipped mobile phone, etc.) that reads the image printed in the step of printing the image is less than the readable area. Thus, the image may be divided (claim 6).

  In addition to the step of printing the image, or instead of the step of printing the image, a step (S104) of displaying on the screen an image in which information for specifying a position on the image or embedded information is embedded is displayed. (Claim 7).

  In the case of including the step of displaying the image on the screen, the method can further include a step (S200) of dividing the image in which the information for specifying the position on the image or the embedded information is embedded (S200). The step of dividing the image is a step of displaying the image on the screen, and the image may be divided so as to be less than an area that can be output on one screen. According to this method, even if the captured image has a low resolution, a high resolution / high definition image can be formed by combining a plurality of captured images.

  Also, in the process of dividing the image, it can be read by one reading of a reading device (for example, a scanner, a digital camera, a camera-equipped mobile phone, etc.) that reads the image displayed in the process of displaying the image on the screen. The image may be divided so as to be equal to or smaller than the area (claim 10).

  Furthermore, it is also possible to include a step of displaying the divided images while switching the images automatically in a constant cycle.

  The information for specifying the position on the image or the embedded information embedded in the input image may be embedded in, for example, a minute region of the input image (claim 12). Further, in this case, the minute region is a rectangular region having a lattice shape at equal intervals, and the position on the input image can be specified from the lattice point coordinates of each rectangular region. .

  Information for specifying the position on the image may be embedded in the minute area as image size and position information. Further, in the minute area, a pattern for identifying the minute area may be set at a specific location in the minute area.

  The information to be embedded in the input image may be divided and embedded to the amount of data that can be embedded in the input image. Alternatively, the information to be embedded in the input image may be divided into a data amount that can be embedded in the minute area and embedded in a plurality of minute areas (claim 17). Further, the information to be embedded in the input image may be embedded in a plurality of minute regions by further adding an image size and position information to the divided information (claim 18).

  According to such a method of embedding embedded information, information embedded as a watermark is divided and embedded in small area units, and each divided information includes position information indicating which part of the entire information is included. By embedding information necessary for composing complete information, it is possible to acquire watermark information embedded with the minimum necessary photographing. In this way, when various information such as copyright information is embedded at the same time as the position information, a large amount of embedded information can be extracted with the minimum necessary shooting, saving labor and time for shooting.

  The information to be embedded in the input image may be embedded in a plurality of minute regions by further adding an identification number to the divided information (claim 19). By embedding the image size and position information together with the identification number, it is possible to detect a captured area / an unphotographed area.

  The size of the rectangular area may be changeable according to the data amount of the embedded information (claim 20). In this way, an area for embedding embedded information can be secured.

  In order to solve the above problem, according to a second aspect of the present invention, there is provided an information extraction method for extracting information embedded in an image, the step of inputting an image (S105), and from the input image, A step of reading information for specifying a position on the image (S106), a step of placing the input image at a predetermined position based on the information for specifying a position on the image (S107), An information extraction method is provided, characterized in that it includes (claim 21).

  In the step of inputting an image, the printed matter according to claims 3 to 6 or the displayed image according to claims 7 to 11 may be input (claim 22).

  In the step of inputting an image, an image that is displayed while automatically switching images sequentially at a constant cycle according to claim 11 may be read as a moving image, and an image of each frame of the moving image may be an input image ( Claim 23).

  Furthermore, a step of reading embedded information embedded in the input image may be included (claim 24).

  The step of reading the information for specifying the position on the image may detect a rectangular rectangular region at equal intervals by detecting a rectangular region identification pattern from the input image. 25).

  The embedded information may be extracted in a state of being divided from the rectangular area, and may be restored to the information at the time of embedding by combining the divided information.

  According to such a method for extracting embedded information, information embedded as a watermark is divided and embedded in a small area unit, and each divided information includes position information indicating which part of the entire information, etc. By embedding information necessary for composing complete information, it is possible to acquire watermark information embedded with the minimum necessary photographing. In this way, when various information such as copyright information is embedded at the same time as the position information, a large amount of embedded information can be extracted with the minimum necessary shooting, saving labor and time for shooting.

  The embedded information is further provided with an identification number, and the presence or absence of the extracted information may be determined by confirming the extracted identification number (claim 27). By embedding the image size and the position information together with the identification number, it is possible to detect a region where information is extracted / a region where information is not extracted.

  When determining whether there is a shortage of extracted information, if it is determined that there is a shortage of extracted information, based on the image size and position information extracted from the same rectangular area, The position may be determined (claim 28). Alternatively, when it is determined that the extracted information is insufficient, the information embedded in the input image may be extracted by additionally inputting an image of a rectangular area in which the insufficient information is embedded (claim 29). .

  According to another aspect of the present invention, there are provided a program for causing a computer to execute the information embedding method or the information extracting method, and a computer-readable recording medium on which the program is recorded. Here, the program may be described in any programming language. In addition, as a recording medium, for example, a recording medium that is currently used as a recording medium capable of recording a program, such as a CD-ROM, a DVD-ROM, or a flexible disk, or any recording medium that is used in the future should be adopted. Can do.

  In the above description, reference numerals in parentheses associated with components and processes indicate corresponding components and processes (steps) in the following embodiments and drawings as an example for easy understanding. However, the present invention is not limited to this.

  As described above, according to the present invention, in a plurality of photographed images, even if there are images that are not connected to the photographing region, or there are omissions / overlaps in the photographed images, the combined images are accurately arranged. Obtainable. Even when the object to be photographed is cut, the arrangement before cutting can be reproduced with an accurate arrangement. In addition, even when an object to be imaged is partially cut and diverted, it can be tracked.

  Further, even if the captured image has a low resolution, a high resolution / high definition image can be formed by combining a plurality of captured images. In addition, by embedding the image size and position information, it is possible to detect a photographed area / an area not photographed. In addition, when various information such as copyright information is embedded at the same time as the position information, the embedded large-capacity information can be extracted with the minimum necessary photographing, and the labor and time of photographing can be saved.

  In the following description of the embodiment, Japanese Patent Laid-Open No. 2003-101762 “Watermark Information Embedding Device and Watermark Information Detection Device” (Patent Document 1), Japanese Patent Laid-Open No. 2003-209676 “Watermark Information Embedding Device, Watermark Information” "Detection device, watermark information embedding device, and watermark information detection device" (Patent Document 3) is a prior invention. First, this prior invention will be outlined.

  The watermark information embedding device described in Patent Document 1 includes a watermarked image synthesizing unit that creates a watermarked image by superimposing a dot pattern on an image. There are a plurality of types of dot patterns, which are regularly arranged, and at least one of them. The kind of dot pattern is characterized in that confidential information representing a specific confidentiality is given. The watermark information detection apparatus described in Patent Document 1 includes a watermark detection unit that detects a watermark image from a watermarked image created by superimposing an image and a watermark image in which a plurality of types of dot patterns are embedded. The watermark detection unit includes a filter that extracts a plurality of types of dot patterns that are the same as the watermark image, and detects the watermark image by matching the watermarked image.

According to the invention described in Patent Document 1, the following effects can be obtained.
(A) Since the embedding information is expressed by the difference in dot arrangement, there is no change in the original document font, character spacing or line pitch.
(B) Since the density (number of dots in a certain interval) of the dot pattern to which the symbol is assigned and the dot pattern to which the symbol is not assigned is equal, the background of the document is shaded with a uniform density to the human eye The information is not conspicuous.
(C) Since the dot pattern to which the symbol is assigned and the dot pattern to which the symbol is not assigned are kept secret, it becomes difficult to decode the embedded information.
(D) A pattern representing information is a collection of fine dots, which are embedded as a background of a document, so that even if an embedding algorithm is disclosed, it is difficult to tamper with embedded information for a printed document.
(E) Since the embedded signal is detected due to the difference in the direction of the wave (shading change) (since detailed detection is not performed in units of one pixel), stable information detection is possible even if there is some dirt on the printed document. It can be performed.
(F) Since the same information is repeatedly embedded and information is restored by using all of the repeatedly embedded information at the time of detection, the signal portion is hidden by large font characters or the paper is dirty. Even if a shortage of information occurs, the embedded information can be taken out stably.

  The invention described in Patent Document 3 calculates an embedding area for embedding a dot pattern in which a predetermined symbol can be identified by a predetermined filter in a document image, and the ratio of the character area to the embedding area is predetermined. A dot that can be used to identify a symbol that forms at least a part of confidential information in an area that does not overlap the character area of the embedding area when the ratio of the character area is equal to or less than a predetermined threshold. A predetermined number of patterns (symbol units) are embedded.

According to the invention described in Patent Document 3, the following effects can be obtained.
(A) By referring to the character arrangement state of the document image into which the electronic watermark is inserted and embedding a valid symbol unit that forms a part of confidential information only in an area other than the character area, what kind of original document is Confidential information can be embedded without fail.
(B) By embedding a predetermined number of symbol units in a region where symbol units are not embedded, it is possible to reliably determine that a valid symbol unit forming part of confidential information is not embedded at the time of detection. it can.
(C) When detecting embedded information, symbols are determined based on the sum of output values of a plurality of types of filters for a certain area, so that the accuracy of information detection is kept high.

  Exemplary embodiments of an information embedding method and an information extracting method according to the present invention will be described below in detail with reference to the accompanying drawings. In the present specification and drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant description is omitted.

(First embodiment)
The first embodiment embeds position information in the background of a printed matter or screen display by using a watermark embedding technique in the printed matter, and based on the position information obtained from the watermarked photographed image, an image that is not connected to the photographing region. Even when there are omissions / overlaps in each other or in the captured images, a plurality of images can be combined in an accurate arrangement.

FIG. 1 is an explanatory diagram showing the configuration of the present embodiment.
As shown in FIG. 1, the present embodiment includes a watermarked image output unit 100 and a watermarked image combining unit 106.

(Watermark image output unit 100)
The watermark image output unit 100 is a device that outputs a watermark image with position information using a digital watermark technique for printed matter. As shown in FIG. 1, the watermark image output unit 100 is a device that receives image data 101, and includes a watermark image forming unit 102, a watermarked image composition unit 103, and an output unit 104. Yes.

  The image data 101 is an image in which a watermark is embedded. The watermark image forming unit 102 is a functional unit (processing or device) that forms a watermark image based on the input image data size. In the watermark image formed in this embodiment, position information capable of specifying the position on the watermark image is embedded (this will be described later with reference to FIG. 3). The watermarked document image composition unit 103 is a functional unit that superimposes image data and a watermark image. The output unit 104 is a functional unit that prints out or displays a watermarked image.

  The watermarked image output 105 is a printed image of a watermarked image printed out by the watermark image output unit 100 or an image displayed on the screen.

(Watermarked image combining unit 106)
The watermarked image combining unit 106 is an apparatus that extracts and combines position information from partially read watermarked images. As shown in FIG. 1, the watermarked image combining unit 106 is a device that receives the watermarked image output 105 as an input, and includes an input unit 107, a watermark detection unit 108, and an image combining unit 109. Yes.

  The input unit 107 is a functional unit that reads a printed matter obtained by printing out a watermarked image or an image displayed on a screen, and specifically includes a scanner, a digital camera, a mobile phone with a camera, and the like. The input unit 107 captures a partial area and acquires a more detailed image (this will be described later with reference to FIGS. 4A and 4B). The watermark detection unit 108 is a functional unit that extracts information from the image read by the input unit 107. In the present embodiment, position information that can specify the position on the watermark image is detected (this will be described later with reference to FIG. 3). The image combining unit 109 is a functional unit that arranges the read image and restores the image based on the detected position information (this will be described later with reference to FIGS. 4A and 4B).

The present embodiment is configured as described above.
Next, the operation of this embodiment will be described in detail with reference to the processing flow of FIG.

<Image data input (S101)>
The image data 101 is input to the watermarked image output unit.

<Watermark image formation (S102, watermark image forming unit 102)>
A watermark image is formed based on the input image data size. As shown in FIGS. 3A and 3B, it is assumed that the watermark images formed in this embodiment are arrayed with watermark images from which information can be extracted in units of equally spaced grid-like rectangular areas. In FIG. 3A, information 0 (Bit 0) and information 1 (Bit 1) are expressed by 4 pixels × 4 pixels. Furthermore, the width / height (w × h) of the image data to be embedded and the position information (2, 1) of the watermark image in the rectangular unit are embedded in each watermark image in the rectangular unit. Here, the image size is expressed by the number of rectangles in the vertical and horizontal directions of the watermark image in the rectangular unit, and the position information is the upper left position (x in the rectangular unit with the upper left of the image size as the origin (0, 0). , Y) are expressed so as to be specified.

  As for the method of detecting a rectangular watermark image, as shown in FIG. 3A, the upper left 3 × 3 signal pattern of the rectangular area is set as a specific pattern for rectangular detection, and the image is partially photographed. It is possible to detect the reading start position in rectangular units. Then, the image size and the image position can be acquired even from a rectangular watermark image partially captured as shown in FIGS. 4A (a) and (b). Further, by embedding copyright information and falsification detection information related to the printed matter and the screen display at the same time as the position information, processing such as copyright information confirmation and falsification detection can be performed.

<Watermarked image composition (S103, watermarked image composition unit 103)>
The image data and the watermark image are superimposed to generate a watermarked image.

<Watermarked image output (S104, output unit 104)>
Print or display the watermarked image on the screen.

<Read (S105, input unit 107)>
As shown in FIGS. 4A and 4B, a printed matter on which a watermarked image is printed or an image displayed on the screen is partially photographed with a camera-equipped mobile phone or the like. At this time, it is sufficient that at least one entire surface of the rectangular watermark image is captured. Further, the shooting order is arbitrary, and it is not always necessary to connect the shooting areas.

<Watermark Detection (S106, Watermark Detection Unit 108)>
As shown in FIG. 4A, an information extraction start point is detected from a partially read image by detecting a rectangular detection pattern embedded in the upper left 3 × 3 signal pattern of the rectangular area, and each rectangular area is detected. Extract image size and position information embedded in units.

  FIG. 4A (a) shows a case where the signal start position is detected and the image size 20 × 30 and the arrangement coordinates (0, 0) are extracted in a rectangular area unit from which information can be extracted. FIG. 4A (b) shows a case where the signal start position is detected and the image size 20 × 30 and the arrangement coordinates (15, 10) are extracted in a rectangular area unit from which information can be extracted.

<Image Combining (S107, Image Combining Unit 109)>
Then, the image is restored as shown in FIG. 4A (c) by arranging the read image at the corresponding coordinates from the watermark image size and position information obtained by the watermark extraction. At this time, if there is an overlap between the captured images, for example, as shown in FIG. Further, it is not always necessary to read the entire surface, and a combined image can be displayed, for example, as shown in FIG. .

  Furthermore, when applied to a copying machine as shown in FIGS. 5A to 5C, the obtained combined image can be output as a copy print. Also, when a printed material divided due to tearing or the like is read by a copy machine or the like, it is sufficient that at least one whole surface of a rectangular watermark image is taken, and the arrangement for reading is arbitrary.

  FIG. 5A shows processing when a watermarked document is divided due to tearing or the like. Even if the watermarked document in (a) is divided due to tearing or the like as in (b), the coordinates are extracted as in (d) by reading as in (c). Can do. Then, image combination is performed as in (e), and printing is performed as in (f), so that the combined image is copied and output as in (g).

  FIG. 5B shows details of the reading of FIG. 5A (c). As in (a), the arrangement at the time of reading is arbitrary. Further, even when there are missing areas due to breakage or the like in the divided areas as shown in (b), they can be combined as shown in (c).

  FIG. 5C shows details of the coordinate extraction of FIG. 5A (c) and the image combination of FIG. 5A (d). FIG. 5C (a) shows details of the coordinate extraction of FIG. 5A (c), and shows a case where the signal start position is detected and the image size 20 × 30 and the arrangement coordinates (0, 0) are extracted. . FIG. 5C (b) shows the details of the image combination of FIG. 5A (d). The arrangement coordinates are extracted by extracting the arrangement coordinates (10, 0) and the images are arranged at the coordinates (10, 0). Is done.

(Effects of the first embodiment)
As described above, according to the present embodiment, the position information is embedded in the background of the printed matter or the screen display, so that the images that are not connected to the shooting area or the shot images are missing or overlapped. However, a combined image can be obtained with an accurate arrangement. In addition, when applied to a copying machine, it is possible to combine the read images of the printed matter divided due to tearing or the like and output the combined copy print.

(Second Embodiment)
In the second embodiment, the watermarked image in which position information is embedded in the first embodiment is matched to the size that can be output by the output unit 104, and the watermarked image is divided and output. A high-resolution and high-definition combined image can be acquired by combining divided images even from a low-resolution display display or a captured image of low-resolution printing such as FAX.

  As shown in FIG. 6, the present embodiment is configured to further include an image dividing unit 200 with respect to the configuration of the first embodiment (FIG. 1). Since the configuration other than the image dividing unit 200 is substantially the same as that of the first embodiment, a duplicate description is omitted.

  The image dividing unit 200 is a functional unit that acquires the output resolution and the maximum outputable size from the output unit 104 and divides the watermarked image into sizes that can be output by the output unit 104.

The present embodiment is configured as described above.
Next, the operation of this embodiment will be described in detail with reference to the processing flow of FIG.

  The image data input step (S101) to the watermarked image composition step (S103) are substantially the same as those in the first embodiment.

<Watermarked image division (S200, image division unit 200)>
The output resolution and the maximum outputable size are acquired from the output unit 104. Then, as shown in FIG. 8, the watermarked image of (a) is divided into sizes that can be output by the output unit 104 as shown in (b). As a result, even when display output with a resolution lower than that of printing or low resolution printing such as FAX is performed, the image enlarged by the reduction in resolution can be divided and a watermarked image can be output. Become.

<Image output with watermark (S201, output unit 104)>
The divided watermarked image is output. As a method of outputting the divided images on the display, as shown in FIG. 9, the divided images are automatically switched in order at a certain cycle ((a) → (b) →... → (c) → (d)), display on the display. In the case of printout, a watermarked printed matter corresponding to the number of divided images is output.

<Read (S202, input unit 107)>
As shown in FIG. 9, an image displayed on a display of a watermarked image output in a divided manner or a printed matter is partially photographed with a camera-equipped mobile phone or the like. By capturing all the divided images, a high-resolution and high-resolution combined image can be acquired even from images that are divided and output by low-resolution printing such as display display with low output resolution or FAX. Become.

  The watermark detection step (S106) to the image combination step (S107) are substantially the same as those in the first embodiment.

(Effect of 2nd Embodiment)
As described above, according to the present embodiment, a watermarked image is divided and output in accordance with the maximum size that can be output by the output unit 104, so that a display with a low output resolution or a low resolution such as FAX can be obtained. A combined image with high resolution and high definition can be acquired by combining the divided images even from a photographed image of printing.

(Third embodiment)
In the third embodiment, the watermarked image in which the position information is embedded in the first embodiment is added to the watermarked image according to the size of the shooting area of the input unit 107 in the same manner as in the second embodiment. A combined image can be acquired without moving the shooting area by continuously shooting the divided divided image display area in which the image is divided and the image is switched at a constant cycle.

  This embodiment is substantially the same as the configuration of the second embodiment (FIG. 6), but the function of the image dividing unit 200 is different. The image dividing unit 200 according to the present embodiment is a functional unit that acquires the output resolution from the output unit 104 and divides the watermarked image so that the image is output to the shooting area size of the input unit 107. Since the configuration other than the image dividing unit 200 is substantially the same as that of the second embodiment, a duplicate description is omitted.

The present embodiment is configured as described above.
Next, the operation of the present embodiment will be described only for steps different from those of the second embodiment (FIG. 7).

<Watermarked image division (S200, image division unit 200)>
The output resolution is acquired from the output unit 104, and as shown in FIG. 10, the watermarked image of (a) is divided so as to be output to the shooting area size of the input unit 107 as shown in (b).

<Image output with watermark (S201, output unit 104)>
The divided watermarked image is output. As a method of outputting the divided watermarked image, as shown in FIG. 11, the divided watermarked image is automatically switched to a fixed area sequentially in a fixed cycle ((a) → (b) → (c) → (d)), by displaying on the display, continuous shooting or moving image shooting can be performed without moving the shooting area.

<Read (S202, input unit 107)>
As shown in FIG. 11, the fixed area where the divided watermarked images are displayed on the display while automatically switching the images sequentially at a fixed period is continuously photographed or animated with a mobile phone with a camera. As a result, a combined image can be acquired from the divided watermarked image that can be photographed without moving the photographing region.

(Effect of the third embodiment)
As described above, according to the present embodiment, a watermarked image is divided according to the size of the shooting area of the input unit 107, and a divided image display area in which the image is switched at a constant period is shot by moving image shooting. The combined image can be acquired without moving the area.

(Fourth embodiment)
The fourth embodiment is characterized in that a large amount of information can be embedded by embedding / extracting embedded information by dividing it into rectangular units using the method of the first embodiment.

  As shown in FIG. 12, the present embodiment is different from the configuration of the first embodiment (FIG. 1) in that embedded information 400 is input to the watermark image forming unit 102. Further, an information combining unit 409 is provided instead of the image combining unit 109. Other configurations are substantially the same as those of the first embodiment, and thus redundant description is omitted.

  The embedded information 400 is information that is embedded in the input image data together with information on the image size and image position as a watermark. The information combining unit 409 is a functional unit that combines extracted data and restores embedded information.

The present embodiment is configured as described above.
Next, the operation of this embodiment will be described in detail with reference to the processing flow of FIG.

<Image data / embedding information input (S401)>
The image data 101 and the embedded information 400 embedded in the image data are input. As the width and height of the input image data and the number of images increase, the amount of information that can be embedded increases.

<Watermark Image Formation (S402, Watermark Image Forming Unit 102)>
As to the formation of the watermark image, as shown in FIGS. 14A and 14B, in addition to the image size and arrangement coordinates, the watermark image area used in the first to third embodiments can extract information in units of rectangular areas. Furthermore, embedding information is added to form a watermark image.

  As shown in FIG. 14A, for the embedded information to be added shown in (a), a checksum is given in advance as shown in (b), an error correction code is given as shown in (c), and (d As shown in (e), the majority data is generated so that the majority of the data extracted at the time of information extraction can be determined, and the same information is repeatedly embedded as the majority data as shown in (e). The embedded information subjected to these processes is divided and embedded in a watermark image in rectangular units.

  For the watermark image in the rectangular unit, as shown in FIG. 14B (f), the area set in the first to third embodiments is expanded so that the divided embedded information can be embedded. Is used as an area for embedded data.

  The watermarked image composition step (S103) to reading step (S105) are substantially the same as those in the first embodiment.

<Watermark Detection (S406, Watermark Detection Unit 108)>
As shown in FIGS. 15A and 15B, information embedded in each rectangular area unit is extracted, and the extracted data is aligned based on the position information. In FIG. 15A, partial images are shot (watermark image input) from the entire image with a camera-equipped mobile phone or the like as shown in (a), and embedded data is extracted in units of rectangular areas as shown in (b). As for the alignment of the extracted data, as shown in FIG. 15B, the embedding order is set in advance at the time of embedding, and the extracted data is aligned according to the embedding order set at the time of embedding.

<Information combination (S407, information combination unit 409)>
In FIG. 15B, as shown in (d), the aligned extracted data are combined to restore the embedded information. For restoration of embedded information, as shown in (e), a majority decision is made on the data in which the same information obtained by combining the aligned extracted data is repeatedly embedded. The embedded information obtained as a result is error-corrected. Finally, if the checksum is calculated as shown in (g) and there is no error, it is output as restored embedded information as shown in (h).

  In the present embodiment, as described above, information is embedded for each rectangular area unit. Therefore, as shown in FIG. 15A, a partially photographed image, a partially cut-out newspaper or magazine scrap, etc. Information can be extracted even from an image read by a copier or the like. In addition, the present embodiment can be applied to a copying machine, a scanner, and the like. That is, as shown in FIG. 16, scraps partially cut as shown in (a) are cut out, rearranged as shown in (b), and then read by a copier or scanner as shown in (c). Thus, as shown in (d), it is possible to extract embedded data in units of rectangles, or to extract information by reading a plurality of scraps together.

(Effect of the fourth embodiment)
As described above, according to the present embodiment, using the method of the first embodiment, the embedded information can be divided and embedded in an image from which information can be extracted in units of rectangular areas. This facilitates information embedding divided into a plurality of printed materials and extraction of information from a plurality of read images, thereby enabling large-capacity information embedding / information extraction. On the other hand, information can be extracted from an image obtained by reading a partially photographed image or a partially cut-out newspaper or magazine scrap with a copier or the like. Furthermore, it is possible to cut out scraps partially cut and change the arrangement, or to read information by extracting a plurality of scraps collectively.

(Fifth embodiment)
In the fifth embodiment, in the fourth embodiment, when the embedded information is restored, there is an unphotographed area, and there is insufficient data for information restoration, or the result of majority decision is obtained. If an error occurs as a result of error correction of the embedded information and calculation of the checksum, all areas are returned even if restoration of the embedded information fails by notifying the area that needs to be re-photographed. It is characterized in that it is not necessary to shoot again.

Since the configuration of the present embodiment is substantially the same as that of the fourth embodiment, redundant description is omitted.
Hereinafter, the operation of this embodiment will be described only for the steps different from those of the fourth embodiment.

<Watermark Image Formation (S402, Watermark Image Forming Unit 102)>
In the previous stage of the formation of the watermark image, repeated data for majority decision is generated through checksum assignment and error correction code assignment as in the fourth embodiment (FIGS. 14A (a) to (e)). In the formation of a watermark image, as shown in FIG. 17, an ID number is given to the information divided and embedded for each rectangular unit as shown in (f), and a rectangular area as shown in (g). Further, an ID number is embedded.

<Watermark Detection (S406, Watermark Detection Unit 108)>
<Information combination (S407, information combination unit 409)>
As in the fourth embodiment, in FIG. 15A, partial images are captured (watermark image input) from the entire image with a camera-equipped mobile phone or the like as shown in (a), and a rectangular area unit as shown in (b). Extracting embedded data at. In the present embodiment, a case will be described in which an area that has not been shot / read exists when embedding information is restored.

  As shown in FIG. 18, as shown in (c), when there is an area that has not been shot / read when embedding information is restored, the ID number is checked from the information embedded in units of rectangular areas. As in (d), for an ID number that has not been detected, the corresponding undetected area is detected from the position information and displayed on the screen as a re-photographing / reading area to prompt re-photographing / reading. For example, when the embedded data of ID4 to 9 is not detected, the positions of ID4 to 9 are calculated from the position information of ID3 and ID10 and displayed on the screen. According to the display of the data non-detected area, the area that has not been imaged / read is re-imaged / read as shown in (e), and the undetected embedded data is acquired as shown in (f).

  In addition to the ID number check, the error information is embedded even if an error occurs as a result of error correction of the embedded information obtained as a result of the majority vote and the checksum calculation. The position information of the rectangular area is detected and displayed on the screen as a re-photographing / reading area to prompt re-photographing / reading.

  In addition, the information extraction process may be interrupted when an insufficient area is not added. This will be described with reference to FIGS. 19A and 19B.

  In FIG. 19A, when the watermark area is insufficient at the time of cutting as shown in (a), the cutting (scrap etc.) of (b) is read as shown in (c). As shown in (d), the embedded data is extracted in units of rectangular areas, and the ID number is checked.

  In FIG. 19B, for example, when embedded data of ID4 to 9 is not detected as shown in (e), the positions of ID4 to 9 are calculated from the positional information of ID3 and ID10 as shown in (f), and the copying machine or scanner Etc. At this time, as shown in (g), additional reading of the insufficient area is performed in accordance with the display of the data undetected area, but if the insufficient area is not added, the information extraction processing is interrupted. Then, an ID number check is performed as shown in (h), and after the unembedded ID4 to 9 embedded data is acquired, embedded information extraction processing is performed.

(Effect of 5th Embodiment)
As described above, according to the present embodiment, in the fourth embodiment, when restoring embedded information, there is a region that has not been shot, and data for information restoration is insufficient. Error correction of embedded information obtained as a result of majority vote and checksum calculation results in an error. If an error occurs, restoration of the embedded information failed by notifying the area that needs to be re-photographed. Even in this case, it is not necessary to re-shoot all areas. Thereby, when extraction of a large amount of information extracted from a plurality of read images fails, it is possible to save the burden of re-photographing. In addition, when applied to a copy machine or the like, copying can be prohibited if copy permission information is not extracted.

  The preferred embodiments of the information embedding method and the information extracting method according to the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to such examples. It will be obvious to those skilled in the art that various changes or modifications can be conceived within the scope of the technical idea described in the claims, and these are naturally within the technical scope of the present invention. It is understood that it belongs.

  The present invention can be used for an information embedding method and an information extracting method, and in particular, can be used for an information embedding method and an information extracting method characterized by image combination by image processing.

It is explanatory drawing which shows the structure of 1st Embodiment. It is explanatory drawing which shows the processing flow of 1st Embodiment. It is explanatory drawing of a watermark image. It is explanatory drawing which shows the process of a reading process-an image combination process. It is explanatory drawing when there exists an overlap or omission in an imaging region. It is explanatory drawing which shows a process when a watermarked document is divided | segmented. It is explanatory drawing which shows the detail of the reading of FIG. 5A. It is explanatory drawing which shows the detail of the coordinate extraction and image coupling | bonding of FIG. 5A. It is explanatory drawing which shows the structure of 2nd Embodiment. It is explanatory drawing which shows the processing flow of 2nd Embodiment. It is explanatory drawing which shows the process of a watermarked image division | segmentation process. It is explanatory drawing which shows the case where the divided | segmented image is output by a display display. It is explanatory drawing which shows the process of a watermarked image division | segmentation process. It is explanatory drawing which shows the case where the divided | segmented image is output by a display display. It is explanatory drawing which shows the structure of 4th Embodiment. It is explanatory drawing which shows the processing flow of 4th Embodiment. It is explanatory drawing which shows the case where the watermark image which added embedding information is formed. It is explanatory drawing which shows the case where the watermark image which added embedding information is formed. It is explanatory drawing which shows the case where a partial image photography is performed with the mobile phone with a camera etc. from a whole image. It is explanatory drawing which shows restoration | restoration of embedding information. It is explanatory drawing which shows the case where 4th Embodiment is applied to a copier etc. FIG. It is explanatory drawing which shows the case where the watermark image which added ID number is formed. It is explanatory drawing which shows the case where embedding data is not detected. It is explanatory drawing which shows the case where 5th Embodiment is applied to a copier etc. FIG. It is explanatory drawing which shows the case where embedding data is not detected.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Watermark image output part 101 Image data 102 Watermark image formation part 103 Watermarked document image composition part 104 Output part 105 Watermarked image output 106 Watermarked image combination part 107 Input part 108 Watermark detection part 109 Image combining part 200 Image division part 400 Embedded information 409 Information coupling part

Claims (29)

An information embedding method for embedding information in an image using a digital watermark,
Inputting the image;
Inputting embedded information to be embedded in the input image;
Repetitively combining the embedded information to generate majority data for repetition,
Dividing the repeated data for majority decision to generate a plurality of embedded data;
A step of sequentially embedding the plurality of embedded data in each area constituting the input image;
An information embedding method characterized by including: In the step of sequentially embedding the plurality of embedded data,
  2. The information embedding method according to claim 1, wherein information for specifying a position on an image is assigned to each of the plurality of embedding data and embedded in each of the regions. Further characterized in that it comprises a step of printing an image in which the plurality of embedded data is embedded in sequence in the method embedding information according to claim 1 or 2. 4. The information embedding method according to claim 3, further comprising a step of dividing an image in which the plurality of embedded data are sequentially embedded . 5. The information embedding according to claim 4, wherein in the dividing step , an output resolution and a maximum outputable size are obtained, and an image enlarged by the reduction in resolution is divided so as to be equal to or less than the maximum size. Method. 5. The information embedding method according to claim 4, wherein in the dividing step, an output resolution and a shooting area size are acquired, and an image enlarged by the reduction in resolution is divided so as to be equal to or smaller than the shooting area size. . The information embedding method according to claim 1, further comprising a step of displaying on the screen an image in which the plurality of embedded data are sequentially embedded . The information embedding method according to claim 7, further comprising a step of dividing an image in which the plurality of embedded data are sequentially embedded . In the step of generating the plurality of embedded data, in the step of displaying the image on the screen, the image in which the plurality of embedded data is sequentially embedded is divided so that the area can be output on one screen or less. The information embedding method according to claim 8, wherein: In the dividing step, the plurality of embedding data are sequentially embedded so as to be less than the readable area by one reading of the reading device that reads the image displayed in the step of displaying the image on the screen. characterized by dividing an image, the method embedded information according to claim 8.   The information embedding method according to any one of claims 8 to 10, further comprising a step of displaying the divided images automatically while sequentially switching the images at a constant period. 2. The information embedding method according to claim 1, wherein the plurality of embedding data embedded in the input image are embedded in a minute region of the input image.   13. The minute area is a rectangular area having a lattice shape at equal intervals, and a position on the input image can be specified from a lattice point coordinate of each rectangular area. Information embedding method. 14. The information for specifying the position on the image is embedded as an image size and position information in a minute area of the input image, according to any one of claims 2, 12, and 13. Information embedding method.   15. The information embedding method according to claim 2, wherein a pattern for identifying a minute area is set in the minute area at a specific location in the minute area. The repeated data majority is characterized to be embedded is divided up to the amount of data that can be embedded in the input image, the method embedding information according to claim 1. 16. The information according to claim 2, wherein the repeated data for majority decision is divided into a data amount that can be embedded in the minute area and embedded in the plurality of minute areas. Embed method. The repeated data for majority decision is further provided with an image size and position information added to the divided information, and embedded in a plurality of the minute regions. Information embedding method. 19. The information embedding method according to claim 17 or 18, wherein the majority data is embedded in a plurality of minute regions by further adding an identification number to the divided information.   14. The information embedding method according to claim 2, wherein a size of the minute area can be changed according to a data amount of the embedding information. An information extraction method for extracting information embedded in an image,
Inputting the image;
Sequentially reading a plurality of embedded data from the input image;
Combining the plurality of embedded data to restore the majority repeated data;
Performing error correction of the embedded information by majority determination of a plurality of embedded information constituting the repeated data for majority determination;
An information extraction method characterized by including:   The input of the image, the printed matter according to claim 3 to 6 or the displayed image according to claim 7 to 11 is input. Information extraction method.   12. In the step of inputting the image, an image that is displayed while automatically switching images sequentially at a fixed period as described in claim 11 is read as a moving image, and an image of each frame of the moving image is read as the input image. The information extraction method according to claim 21, wherein: further,
Reading information for specifying a position on the image from the input image;
Arranging the input image at a predetermined position based on information for specifying a position on the image;
The information extraction method according to claim 21, comprising:   The step of reading information for specifying a position on the image is characterized by detecting a lattice-shaped rectangular region at equal intervals by detecting a minute region identification pattern from the input image. The information extraction method according to claim 24.   26. The information extracting method according to claim 25, wherein the embedded information is extracted in a state of being divided from the rectangular area, and can be restored to information at the time of embedding by combining the divided information.   27. The information extraction method according to claim 26, wherein an identification number is further given to the embedded information, and whether or not the extracted information is insufficient is determined by checking the extracted identification number.   When it is determined that the extracted information is insufficient, the position of the rectangular area in which the insufficient information is embedded is determined based on the image size and position information extracted from the same rectangular area. 27. The information extraction method according to 27.   The information embedded in the input image is extracted by additionally inputting an image of a rectangular area in which the insufficient information is embedded when it is determined that the extracted information is insufficient. Information extraction method.

Images