JPH11196259A - Information transmission method and system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an improved information transmission method by which information is properly extracted and deterioration in the image quality is suppressed in the case of segmenting a document image. SOLUTION: A transmitter side divides an electronic image into division images each stored in a partial area and each division image is classified into a text area or an image area in response to an area characteristics. Then after imbedding copyright information in a form in response to each area characteristics into division images A, B, D of the text area and a division image C of the image area respectively, the division images are assembled into a form of the document image to obtain the entire image. At a receiver side, the received entire image is subject to area divisions similar to the case with the information imbedding, and the imbedded copyright information is read from the division images A, B, D of the text area and the division image C of the image area respectively.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information transmission method and an information transmission system in which specific information of copyright information is embedded and transmitted in a form invisible to an electronic image to be transmitted.

[0002]

2. Description of the Related Art Generally, electronic images can be easily copied using a computer, and there is a high possibility that copyrights will be infringed by transmitting the copied electronic images through a network. Therefore, when transmitting an electronic image through an open network environment such as the Internet, for example, copyright information such as the name of a valid author is embedded in the electronic image in a form invisible to human eyes.
By reading this copyright information when necessary,
A data hiding method for confirming whether or not the electronic image is authentic has been proposed.

[0003] By using this data hiding technique, the receiver can easily confirm the copyright information added to the electronic image transmitted through the network.
There is an advantage that the copyright need not be displayed on the electronic image and the electronic image can be kept as the original electronic image.

As a well-known technique related to a method of embedding information in an electronic image, a technique showing a method of expanding an electronic image in a frequency domain and processing a specific frequency component (see "Nakamura, Ogawa, Takashima") : Digital Watermarking in the Frequency Domain for Copyright Protection of Digital Electronic Images, The 1997 Symposium on Cryptography and Informa
(Technology disclosed in “Action Security-26A”) and a technique that directly processes the density value of each pixel constituting an electronic image (Shizumizu, Numao, Morimoto: Still electronic image data using pixel blocks) Hiding, the technology disclosed in the 53rd National Convention of IPSJ 2-257), and the like.

[0005]

However, among the conventional data hiding methods, for example, a method of processing a specific frequency component, the transmission side treats the entire electronic image as one data unit and transmits the copyright information. For embedding, the entire electronic image is required to extract the copyright information on the receiving side. However, when the electronic image is a document image, if only local parts such as figures, graphs, tables, and paragraphs in the document image are cut out without permission, copyright information cannot be extracted from the cut out image, or The probability of not being able to withdraw is extremely high. In this case, for example, the copyright owner of the original document image may not be known, and may be illegally copied or falsified without the consent of the copyright owner. It cannot be prevented.

As a measure to avoid such a problem, it is conceivable to divide an electronic image into a plurality of pixel blocks having a uniform area, and embed copyright information by regarding each of the pixel blocks as a data unit. . However, in the case of this method, there is no problem if the copyright information is embedded in the cut-out pixel block, but the copyright information cannot be extracted if the copyright information is not embedded. Also, since the copyright information is embedded by dividing it into pixel blocks having a uniform area on the transmission side, the image quality is degraded or the amount of embedded information is restricted. For example, since copyright information is embedded in a pixel block consisting of almost white pixels, in such a case, the image quality is remarkably deteriorated, or a large amount of information cannot be embedded.

Therefore, an object of the present invention is to prevent information embedded in an electronic image from being lost on the transmitting side even if an operation of cutting out the electronic image is performed on the receiving side, and to improve deterioration of the image quality of the electronic image. It is to provide an improved information transmission method. Another object of the present invention is to provide an information transmission system suitable for implementing the above method.

[0008]

An information transmission method according to the present invention for solving the above-mentioned problems includes a first step of embedding and transmitting specific information in a predetermined electronic image in a form that cannot be visually recognized, and a step of transmitting the specified information from the transmitted electronic image. Reading the embedded specific information. The first step is to classify the divided image for each partial area obtained by dividing the electronic image into any one of characteristic areas such as a text area and an image area according to the characteristic of the partial area, and perform the classification. After embedding the specific information in each of the divided images of the characteristic region in a form corresponding to the characteristic of the characteristic region, reassembling the divided images into the original electronic image form to reconstruct the entire image, In the two steps, the divided images obtained by dividing the whole image for each partial region are classified into any of characteristic regions such as a text region and an image region according to the characteristics of the partial region. A step of reading the specific information embedded in the divided image of the area in accordance with a rule at the time of the embedding.

[0009] An information transmission system for solving the above-mentioned other problems is configured to include an information embedding device and an information reading device. The information embedding device is a device that creates an electronic image addressed to a receiving device including a unit that reads out the specific information from the electronic image in which the specific information is embedded, in a form corresponding to the embedding form. Area dividing means for dividing the divided images into partial images for each partial area; classification means for classifying each divided image into one of characteristic areas such as a text area and an image area according to the characteristics of the area; Information embedding means for acquiring specific information to be embedded in each of the divided images, and embedding the specific information in the divided image in a form corresponding to the characteristics of the classified characteristic region; And an image reconstructing means for reconstructing the electronic image addressed to the receiving device by assembling the divided image embedded with the original electronic image.

[0010] The information reading apparatus further comprises: an image acquisition unit for acquiring an electronic image in which specific information is embedded in a divided image classified in advance into a characteristic area such as a text area or an image area in accordance with the area characteristic; A region dividing unit that divides the electronic image into divided images for each partial region, a classifying unit that classifies each divided image into a characteristic region corresponding to the region characteristic, and a characteristic region from the classified divided image. Means for reading out the specific information in a reading form unique to.

[0011]

Embodiments of the present invention will be described below in detail with reference to the drawings. (First Embodiment) First, an embodiment of the information transmission method of the present invention will be described. According to this method, the transmitting side embeds specific information created from an electronic image in an invisible form in the electronic image and transmits the embedded information, and the receiving side reads the specific information from the electronic image. A specific procedure for that will be described with reference to FIG.

On the transmitting side, first, the target electronic image is divided into partial areas (here, four) (FIG. 1A).
The two divided images A to D are extracted (FIG. 1B). Then, each of the divided images A to D is classified into a characteristic region corresponding to the characteristic of the partial region, for example, one of a text region and an image region. The characteristics of the partial region include, for example, a pixel density value, a density distribution, and the number of gradations for expressing an image in the region. In the illustrated example, the divided images A, B, and D are classified into text regions, and the divided image C is classified into image regions (FIG. 1C).

Next, specific information is embedded in each of the divided images in the text area and the image area. Here, after the specific information is embedded in the divided images A, B, and D of the text region and the divided image C of the image region in a form corresponding to the characteristics of the respective characteristic regions (FIG. 1D), each divided image A ~ D
Are assembled into an original electronic image to reconstruct an electronic image (hereinafter, referred to as an entire image for convenience to distinguish it from the original electronic image) (FIG. 1E). The whole image obtained here is transmitted to a receiving side through a network, for example.

On the other hand, on the receiving side, the transmitted whole image is input (FIG. 1 (f)), and this is divided into the divided images A to
D (FIG. 1 (f)), and the divided images A to D are classified into characteristic regions, for example, one of a text region and an image region, according to the characteristics of each partial region. In this example, the divided images A, B, and D are classified into a text area, and the divided image C is classified into an image area (FIG. 1 (g)).

Next, specific information is read from each of the divided images A to D of the text area and the image area in accordance with a reading mode unique to each area, that is, a rule for embedding information in each area (FIG. 1 (h )). As a result, the specific information can be read from any of the divided areas. As the specific information, copyright information of the electronic image, copyright information of each divided image, a summary of the electronic image, a keyword, and the like can be used. In addition, the specific information may be the same for all divided images, or information different in each region, for example, information such as a sequence number may be embedded in all divided images in the characteristic region. good.
In the former case, the specific information can be read from any of the divided images on the receiving side. In the latter case, the receiving side can easily determine which divided image has been cut out without notice by decoding the sequence number and the like.

(Second Embodiment) Next, an information transmission system according to an embodiment of the present invention will be described. This information transmission system is configured by connecting a transmitting device and a receiving device to a network. The transmitting device includes an information embedding device that embeds the specific information in the electronic image, and the receiving device includes an information reading device that reads the specific information.

The information embedding device obtains an electronic image to be transmitted and specific information to be embedded in the electronic image, and divides the divided image of each partial region from the electronic image into a characteristic region corresponding to a region characteristic, For example, it has a function of classifying it into a text area or an image area, embedding specific information in each of the classified divided images, and then reconstructing the entire image.

In order to realize this function, in the present embodiment, as shown in FIG.
2, area dividing section 13, area classifying section 14, area data accumulating section 15, image output section 16, image reconstructing section 17, information input section 18, information accumulating section 19, information embedding section (A, B) 20,
The information embedding device 100 including the information embedding device 21 is configured.
The information embedding device 100 can be realized by a general-purpose computer such as a personal computer or a workstation reading and executing a predetermined program recorded on a recording medium.

Hereinafter, a 256-gradation grayscale bitmap format document image in which one pixel is represented by 8 bits is used as an electronic image, and specific information to be embedded in a divided image obtained by dividing this document image is 8 The operation of the information embedding device 100 will be described assuming that the information is bit copyright information.

First, from the document image stored in a memory (not shown), the one into which copyright information is to be embedded is read out by the image input unit 11, and this document image is stored in the image storage unit 12. Also, the copyright information input by the keyboard or the like is received through the information input unit 18 and stored in the information storage unit 19. The region dividing unit 13 divides the document image stored in the image storage unit 12 into regions such as figures, tables, and paragraphs.

FIGS. 3A and 3B are diagrams for explaining the process of dividing the area of the document image. FIG. 3A relates to a graph of the number of black pixels, and FIG. It is about. Here, for a document image as shown in FIG. 3B, a pixel having a density value equal to or higher than a certain value (for example, 128) is “1” (black pixel), and a pixel having a density value lower than “1” is “0”. "(White pixels) are binarized, and the number of black pixels included in each line is counted for each line, and the vertical axis as shown in FIG. A graph is created in which the coordinate values of the directions and the horizontal axis are the total number of black pixels. At this time, the document image is divided into a plurality of regions at positions where the frequency becomes “0” continuously for 10 lines or more. Further, a similar graph is created by summing up the black pixels of all the columns in the divided area, and similarly, the division is performed again at a position where the frequency becomes “0” continuously for 10 or more columns. Such an operation is repeated until the region is no longer divided.

As a result, it becomes possible to finally extract the divided images corresponding to the seven partial areas as shown in FIG.
These seven divided images and respective position information (the upper left coordinates indicating the relative position of each divided image with respect to the original electronic image, etc.)
Are stored in the area data storage unit 15.

The area classifying section 14 has an area data accumulating section 15
Based on the characteristics of the divided images stored in the image processing apparatus, for example, the pixel distribution, the number of gradations, and the position information, an information embedding method adapted to the characteristics of the partial region is set. The data is stored in the storage unit 15. Here, it is assumed that the image is classified into either a text area or an image area based on the characteristics of each divided image stored in the area data storage unit 15.

When setting the information embedding method in the area classifying unit 14, first, as shown in FIG. 5, a specific divided image (a local part is shown here for the sake of simplicity). Then, the minimum rectangle including the adjacent black pixels in is obtained, and then the ratio of the sum of the rectangular areas to the total area of the divided image is calculated. As a result, if the sum of the rectangular areas is, for example, less than 80% of the total area, the area is classified into a text area, and a tag “T” is attached to the divided image. On the other hand, if the ratio of the sum of the rectangular areas is 80% or more, the image is classified into an image area, and a tag “I” is attached to the divided image.

For example, the ratio of the sum of the rectangular areas in the divided image shown in the upper part of FIG.
%, It is classified into a text area. On the other hand, FIG.
The ratio of the sum of the rectangular areas in the divided image shown in the upper row is
Since it is almost 100% as shown in the lower row, it is classified into an image area. In any case, the divided images with the tags are stored in the area data storage unit 15.

The information embedding unit A20 includes the area data storage unit 1
5 according to the setting result of the information embedding stored in the tag T.
The copyright information stored in the information storage unit 19 is embedded in the divided images marked with. Then, the divided image in which the copyright information is embedded is stored in the area data storage unit 1 again.
Save to 5. Similarly, the information embedding unit B21 converts the copyright information stored in the information storage unit 19 for the divided image with the tag I in accordance with the setting result of the information embedding stored in the area data storage unit 15. Embed. And
The divided image in which the copyright information is embedded is stored in the area data storage unit 15 again.

The function of embedding information will be described more specifically. The information embedding unit A20 embeds copyright information in the divided image of the text area as follows. First, the divided image classified into the text area is binarized, and a minimum rectangle including adjacent black pixels is obtained from the binarized image. Next, an interval between the minimum rectangles in the line direction is calculated. Thereafter, as shown in FIG. 7A, the rectangular pairs on both sides are selected in order from the one with the longest interval, and the copyright information is embedded according to the procedure described below. That is,
The distance between a pair of rectangles whose interval length is m-th is compared with the m-th bit of the 8-bit copyright information to be embedded, and m = 1
The processing corresponding to each of the states (8) to (8) is performed.

For example, the m-th bit of the copyright information is “0”
If the distance between the pair of rectangles is an even-numbered bit, no processing is performed. so,
The distance between the rectangle pairs is set to an even number of bits. FIG. 7 (b)
Is a diagram showing this operation. On the other hand, the copyright information m
If the bit number is “1”, no processing is performed if the distance between the rectangular pairs is an odd number bit. Is replaced by a black pixel, and the distance between the rectangular pairs is set to odd bits.

The information embedding section B21 embeds the copyright information in the divided image of the image area as follows.
Here, for example, the technology disclosed in the above-mentioned known document “Shimizu, Numao, Morimoto: Still Electronic Image Data Hiding Using Pixel Blocks, IPSJ 53rd National Convention 2-257” is applied.

More specifically, first, the divided image of the image area is divided into four equal parts in each of the vertical and horizontal directions to create a total of 16 pixel blocks, and the average value and the variance of the density values of all the pixels in each pixel block are obtained. Find the value. Next, the four pixel blocks as viewed from the horizontal direction are divided into two sets of left and right sides, and eight pixel block pairs are created from a total of 16 pixel blocks. The variance values of the density values of all the pixels are compared. As a result, when the variance value of the left block is equal to or greater than the variance value of the right block, this pixel block pair expresses information of a bit value “0”, while the variance value of the left block is the variance value of the right block. If the value is smaller than the value, it is assumed that this pixel block pair expresses information of a bit value “1”. Here, each bit value information of each pixel block when the pixel block pair is viewed from upper left to lower right is collated with predetermined copyright information to be embedded. As a result, if there is a pixel block pair that does not match the copyright information to be embedded, in each pixel block of the pixel block pair, only the variance value is replaced with the average value as it is by linear conversion.

For example, when the variance value of a specific pixel block A is replaced with the variance value of another pixel block B, the density value N of the new pixel block A is calculated by dividing the average value of the density values of all the pixels in the pixel block A by Ma. Then, it can be obtained by the relationship of “N = Ma + (density value of pixel block A−Ma) × dispersion value of pixel block B / dispersion value of pixel block A”.

The image reconstructing unit 17 reconstructs the entire image from the divided images in which the copyright information is embedded and the information on the respective position coordinates, and stores the reconstructed image in the image storage unit 12. After that, the image output unit 13 outputs the entire image finally stored in the image storage unit 12 as a document image to be transmitted.

The whole image can be stored in a memory (not shown). Further, the specific configuration of the information embedding device 100 shown in FIG. 2 is an example, and another configuration may be used as long as it has the basic function described first.

Next, an information reading device provided in the receiving device will be described. This information reading device divides the whole image transmitted from the transmitting device including the information embedding device 100 into divided images for each partial area, and divides each divided image into a text area, an image area, or the like according to the area characteristic. It has a function of classifying the information into regions and detecting and reading out copyright information embedded in each divided image.

In order to realize this function, in the present embodiment, as shown in FIG.
2. An information reading apparatus 200 including an area dividing section 33, an area classifying section 34, an area data accumulating section 35, an information detecting section A36, an information detecting section B37, an information output section 38, and an information accumulating section 39 is configured. . This information reading device 200
Can be realized by a general-purpose computer such as a personal computer or a workstation reading and executing a predetermined program recorded on a recording medium.

The operation of the information reading device 200 is as follows. First, the entire image is input by the image input unit 31 and stored in the image storage unit 32. Area division unit 3
3 divides the entire image stored in the image storage unit 32 into divided images for each partial area and causes the area data storage unit 35 to store the divided image. In addition, the area classification section 34 includes an area data storage section 35.
The information embedding method adapted to each of the divided images stored in the area data storage unit 35 is set for each divided image based on the characteristics and the like of the individual divided images stored in the area data storage unit 35. The processing up to this point is the same as in the case of the information embedding apparatus 100, and tags T and I are added based on the characteristics of each divided image stored in the area data storage unit 35 to specify either a text area or an image area. Assign to divided images.

In the information detecting section A36, the copyright information is read from the divided image having the tag T, and stored in the information storage section 39. On the other hand, the information detecting unit B37 reads out the copyright information from the divided image with the tag I and stores the copyright information in the information storage unit 39.

After that, the information output unit 38 outputs the information
The copyright information assigned to the divided image in the text area and the copyright information assigned to the divided image in the image area are output. The copyright information output here is filed in a memory (not shown),
Alternatively, it is displayed on a monitor screen of a display device (not shown).

The information reading device 20 shown in FIG.
The specific configuration of 0 is also an example, and another configuration may be used as long as it has the basic function described first. Although the present embodiment is as described above, the present invention is not limited to the above example, and can be implemented in various forms. For example, it is possible to embed and read the copyright information for each divided image of the characteristic region where the cutout operation is easily performed, and to embed and read different copyright information for each divided image. Also, the copyright information to be embedded and read may be other than the copyright information.

[0040]

As is apparent from the above description, according to the present invention, it is possible to embed and read out specific information in units of characteristic regions or in units of divided images, whereby a part of an electronic image is cut out. The necessary information can be read out even if it is read.

Further, by using the specific information as copyright information, unauthorized copying of figures and tables can be suppressed, and when it is discovered that a partially cut out divided image has been copied without permission, Since copyright information can be derived from the divided image, unauthorized reproduction can be suppressed.

Furthermore, since the embedding method can be selected according to the characteristics of the divided image, it is possible to avoid deterioration in image quality due to embedding. Also, by embedding and reading different information for each divided image unit, for example, it becomes possible to embed and read explanation information and the like for each drawing individually.

[Brief description of the drawings]

1 (a) to 1 (h) are explanatory diagrams showing a procedure of an embodiment of an information transmission method according to the present invention.

FIG. 2 is a block diagram showing a configuration example of an information embedding device provided in a transmission device of the information transmission system of the present invention.

FIG. 3 is a schematic diagram for explaining an electronic image area dividing process in the information embedding device described in FIG. 2;
(A) relates to a graph of the number of black pixels, and (b) relates to a document electronic image provided for division.

FIG. 4 is an explanatory diagram of an entire image reconstructed by assembling divided images.

FIG. 5 is a schematic diagram of a minimum rectangle including adjacent black pixels in a divided image shown for explaining the first stage of setting of an information embedding method by the information embedding device described in FIG. 2;

6A and 6B are diagrams for explaining a later stage of setting of an information embedding method by the information embedding device described in FIG. 2, in which FIG.
(B) relates to a classification example of the image area.

FIGS. 7A and 7B are schematic diagrams illustrating the embedding of copyright information in a divided image of each partial area of a text area. FIG.

FIG. 8 is a block diagram showing a configuration example of an information reading device provided in a receiving device of the information transmission system of the present invention.

[Explanation of symbols]

 11, 31 Image input unit 12, 32 Image storage unit 13, 33 Area division unit 14, 34 Area classification unit 15, 35 Area data storage unit 16 Image output unit 17 Image reconstruction unit 18 Information input unit 19, 39 Information storage unit Reference Signs List 20 information embedding part A 21 information embedding part B 36 information detecting part A 37 information detecting part B 38 information output part 100 information embedding device 200 information reading device

Claims (7)

[Claims] A first step of embedding the specific information in a predetermined electronic image in a form that cannot be visually recognized and transmitting the embedded information; and a second step of reading the embedded specific information from the transmitted electronic image. One stage is to divide the electronic image into divided images obtained for each partial region,
Each of the partial regions is classified into any one of characteristic regions such as a text region and an image region according to the characteristics of the partial region, and the specific information is embedded in divided images of the classified characteristic regions in a form corresponding to the characteristics of the characteristic region. After that, a step of assembling these divided images into the form of the original electronic image and reconstructing the whole image is included. The second step is to divide the divided image for each partial region obtained by dividing the whole image,
Each of the partial areas is classified into one of characteristic areas such as a text area and an image area according to the characteristic of the partial area, and the specific information embedded in the divided image of the classified characteristic area is read out according to the rule at the time of the embedding. An information transmission method comprising a step. 2. The information transmission method according to claim 1, wherein in the first step, the same specific information is embedded in all divided images. 3. The method according to claim 1, wherein in the first step, information uniquely representing the content of each of the characteristic regions is embedded in all divided images in the characteristic region.
Information transmission method described. 4. The information transmission method according to claim 1, wherein the specific information to be embedded in the first step is information such as a sequence number which is different for each partial area. 5. An apparatus for creating an electronic image addressed to a receiving device comprising means for reading out the specific information from the electronic image in which the specific information is embedded, in a form corresponding to the embedding form, comprising: Area dividing means for dividing into divided images for each partial area; classifying means for classifying each divided image into one of characteristic areas such as a text area and an image area according to characteristics of the area; Information embedding means for acquiring specific information to be embedded in each of the divided images thus obtained, and embedding the specific information in the divided image in a form corresponding to the characteristics of the classified characteristic region; An information embedding apparatus, comprising: image reconstructing means for reassembling an electronic image addressed to the receiving apparatus by assembling a divided image embedded with an image into an original electronic image. 6. An image acquisition means for acquiring an electronic image in which specific information is embedded in a characteristic region in a divided image previously classified into a characteristic region such as a text region or an image region according to region characteristics, and Area dividing means for dividing the electronic image into divided images for each partial area; classifying means for classifying each divided image into a characteristic area corresponding to the area characteristic; Means for reading out the specific information in the above-mentioned readout form. 7. An information transmission system comprising a transmitting device provided with the information embedding device according to claim 5 and a receiving device provided with the information reading device according to claim 6, connected via a communication line.