JP4626565B2 - Image processing apparatus, image forming apparatus, and program - Google Patents

Description

  The present invention relates to a technique for preventing falsification of a document.

In recent years, the need for information security has only increased, and a technique for preventing falsification of documents and the like is required. For example, Patent Document 1 discloses a copy in which dependency information unique to a document is created from electronic information obtained by reading the document (printed material) using a hash function or the like, and this is added to a blank portion in a barcode format. A technique for generating is disclosed. According to this technique, since the dependency information changes when the content of the document is changed, it is described that the falsification of the document can be easily detected.
JP-A-8-297743

  However, it is very easy to erase such a barcode. For this reason, even if a copy is created using the technique described in Patent Document 1, the dependency information cannot be calculated if the barcode is deleted with a correction liquid or the like, so that tampering can be performed freely. End up. Therefore, the technique described in Patent Document 1 is actually difficult to say as a practical solution for detecting and preventing tampering.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a technique that makes it difficult to conceal falsification and suppress falsification of a document.

In order to achieve the above-described object, the present invention provides an input means for inputting image data represented by a plurality of pixels in an on state or an off state, and a state of each pixel of the image data input by the input means. And a watermark pattern represented by a plurality of pixels smaller than the total number of pixels of the image data based on the feature amount calculated by the feature amount calculation unit and the feature amount calculation unit that calculates the feature amount of the image data A feature amount pattern generating unit that generates a region, a region specifying unit that specifies a plurality of regions having different attributes from the image data input by the input unit, a region specified by the region specifying unit, and an image For each pixel constituting the data, the pixels included in the area where the watermark pattern is embedded and the area where the watermark pattern is not embedded are included. Area information generating means for generating area information which is a numerical value calculated based on the number of pixels to be generated, and a plurality of pixels smaller than the total number of pixels of the image data based on the area information generated by the area information generating means Area information pattern generation means for generating a watermark pattern represented by the state of the image, and a predetermined part or all of the pixels of the area specified by the area specification means of the image data input by the input means, and embedding means for embedding repeating the watermark pattern generated by the watermark pattern generated by the feature amount pattern generating means and the area information pattern generating means, outputs a write Mareta image data embedded watermark pattern by said embedding means Output means that outputs at least one of the image data and the watermark pattern. It was turned on, to provide an image processing apparatus and an output means for outputting a set of pixels that the other pixels in an off state as the image data with the watermark.
According to such an image processing apparatus, a copy is formed in which a watermark pattern generated based on a feature amount representing the state of each pixel of image data is embedded in a part or all of a document. Since this watermark pattern changes according to the feature amount of the image data, it is possible to easily detect falsification of the document. Further, for example, a watermark pattern can be prevented from being embedded in an image area such as a photograph, and a significant deterioration in image quality can be suppressed. Furthermore, it is possible to easily distinguish between a region where a watermark pattern is embedded and a region where it is not.

The present invention also provides an acquisition means for acquiring image data representing an original, an extraction means for extracting a watermark pattern from the image data, and a watermark on the image data acquired by the acquisition means based on an extraction result by the extraction means. A first determination unit that determines whether or not a pattern is included; and when the first determination unit determines that a watermark pattern is included in the image data, A second determination unit that determines whether or not there is a region information pattern that represents region information that is information for specifying a region in which a watermark pattern is to be embedded; and the second determination unit determines that the region information pattern exists. The region information is identified from the region information pattern, and the region identified by the region information and the watermark pattern are extracted. A comparison means for comparing the degree of coincidence with the determined area, and when the degree of coincidence compared by the comparison means is less than a predetermined ratio, the area specified by the area information and the watermark pattern are extracted. An image forming apparatus is provided that includes an image forming unit that forms an image in which a non-coincidence area with the area is filled .

The present invention can also be specified as an image processing method executed by the above-described image processing apparatus and a program for realizing the method. For example, the program is a program for embedding a watermark pattern in image data represented by a plurality of pixels that are in an on state or an off state, from the state of each pixel of the image data input to the computer. a first processing for calculating a feature quantity of the image data, on the basis of the feature quantity calculated by the first processing, the watermark pattern represented by a plurality of the pixels smaller than the total number of pixels of the image data The second process to be generated, the third process for specifying a plurality of areas each having a different attribute from the image data, and the pixels constituting the area specified by the third process and constituting the image data Is calculated based on the number of pixels included in the area where the watermark pattern is embedded and the number of pixels included in the area where the watermark pattern is not embedded. And a watermark pattern represented by a plurality of pixel states smaller than the total number of pixels of the image data based on the region information generated by the fourth processing And the watermark pattern generated by the second process and the first pattern in a predetermined part or all of the pixels of the region specified by the third process of the image data. and the repeating embedding sixth processing watermark pattern generated by the processing of 5, the sixth a processing process of outputting the watermark pattern embedding Mareta image data by the, the image data and the watermark pattern at least one of the pixels in the oN state to the oN state, and a seventh process of outputting the set of pixels to the other pixels in an off state as the image data with the watermark Is a program for row.

  As described above, according to the present invention, it is difficult to conceal falsification, and falsification of a document can be suppressed.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following embodiment, a case where the present invention is applied to a so-called multi-function machine having both a copy function and a print function will be described as an example. For convenience of explanation, in the embodiment described below, the image formed on the document is assumed to be a monochrome binary image.

[Configuration of the embodiment]
FIG. 1 is a block diagram showing a configuration of a multifunction peripheral MFP according to an embodiment of the present invention. As shown in the figure, the configuration of the MFP MFP is roughly divided into an image processing device 10, an image reading device 20, an image forming device 30, and an input / output device 40. The image reading device 20 is a so-called scanner and optically reads a document to generate image data. The image data generated by the image reading device 20 is bitmap data including a plurality of pixels arranged in the main scanning direction and the sub-scanning direction. The image forming apparatus 30 is a so-called printer, and forms an image corresponding to the image data supplied from the image processing apparatus 10 on a sheet-like object such as paper. Here, it is assumed that the image forming apparatus 30 is a monochrome printer. The input / output device 40 includes a touch panel, various buttons, and the like, and receives instructions from an operator and displays various instructions.

  The image processing apparatus 10 performs image processing for preventing falsification on the input image data and supplies the processed image data to the image forming apparatus 30. The image processing apparatus 10 includes a control unit 101, a storage unit 102, a network interface unit 103, a scanner interface unit 104, a printer interface unit 105, and a user interface unit 106.

  The control unit 101 is an arithmetic unit including a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory), and the CPU stores the data in the ROM or the storage unit 102 using the RAM as a work area. By executing the program, the operation of each unit of the image processing apparatus 10 is controlled. Further, the control unit 101 executes image processing on the image data. The storage unit 102 is a non-volatile storage device such as an HDD (Hard Disk Drive), and stores a program executed by the control unit 101, data necessary for image processing, and the like.

  The network interface unit 103 is an interface for connecting to a network NW such as a LAN (Local Area Network), and exchanges image data and the like with the personal computer PC via the network NW. The personal computer PC outputs data described in a page description language (PDL) (hereinafter referred to as “PDL data”) as image data. The scanner interface unit 104, the printer interface unit 105, and the user interface unit 106 are interfaces for connecting to the image reading device 20, the image forming device 30, and the input / output device 40, respectively.

[Operation]
Based on the above configuration, the MFP MFP receives input of image data to be a document via the network NW or the image reading device 20, and executes a copying process for forming an image corresponding to the image data on a sheet. . When executing this copying process, the MFP MFP also executes a process of superimposing and outputting the watermark image on the image data, and a process of reading the watermark image from the image data and detecting falsification. Below, the specific content of these processes is demonstrated.

  The MFP MFP is configured to selectively execute a first mode in which a watermark image is superimposed on the entire image data and a second mode in which the watermark image is superimposed on a partial area of the image data. The processing contents change according to each mode. Therefore, hereinafter, the first mode will be described as an operation example 1, and the second mode will be described as an operation example 2. The first mode is a mode suitable for making it more difficult to falsify the document, and the second mode is a mode suitable for maintaining better readability of the document.

  For convenience of explanation, it is assumed that the image data generated by the image reading device 20 is binary bitmap data. That is, the image data is composed of pixels that signify printing and pixels that signify no printing. Hereinafter, the state of a pixel that means printing is referred to as an “on state”, and a pixel that is in an on state is referred to as an “on pixel”. In addition, a pixel state that means not printing is referred to as an “off state”, and a pixel in an off state is referred to as an “off pixel”.

[Operation Example 1]
FIG. 2 is a flowchart showing a process in which the image processing apparatus 10 of the MFP MFP generates image data with a watermark image in the first mode. Referring to the figure, first, the control unit 101 of the image processing apparatus 10 receives (that is, inputs) image data via the network interface unit 103 or the scanner interface unit 104 (S1101). Next, the control unit 101 determines whether the input image data is bitmap data or PDL data (S1102). If the image data is PDL data, the control unit 101 executes rasterization processing on the image data and generates binary bitmap data corresponding to the PDL data (S1103). If the image data is bitmap data, the control unit 101 skips the process of step S1103.

  Subsequently, the control unit 101 divides the entire bitmap data into blocks each having a predetermined number of pixels (S1104). Although the block size is arbitrary, for example, 9 pixels (3 × 3 pixels), 16 pixels (4 × 4 pixels), and the like are suitable. When the bitmap data is divided, the control unit 101 determines whether the state of each block is “ON” or “OFF” (S1105).

  Here, an example of a determination method performed by the control unit 101 will be specifically described. The control unit 101 refers to the pixels included in the block to be determined, and counts the number of on-state pixels and off-state pixels. Then, the control unit 101 determines that the block is in the on state if the on-state pixels are more than half of the whole, and determines that the block is in the off state if the on-state pixels are less than half of the whole. To do. The control unit 101 executes such determination for all the blocks included in the bitmap data, and specifies the state of each block.

  The determination method is not limited to the above-described example, and any method may be used as long as it is a predetermined procedure for obtaining a unique result. For example, when the number of pixels in the on state is 2/3 or more of the whole, it may be determined that the pixel is in the on state, or the state may be determined based on the number of pixels in the off state.

  When the determination is made for each block, the control unit 101 calculates the feature amount of the bitmap data based on the determination result (S1106). The feature amount of the bitmap data may be calculated using a method such as a hash function or a checksum, or may be calculated using a frequency component obtained by frequency-decomposing the bitmap data. The feature amount of the bitmap data changes when the state (on / off) of the blocks constituting the bitmap data changes, but does not necessarily change only by the change of the pixel state. By doing in this way, the fluctuation | variation of the feature-value by the slight change of a pixel like noise can be suppressed, and a feature-value can be made strong to noise.

  After calculating the feature amount of the bitmap data, the control unit 101 generates a watermark pattern using this feature amount (S1107). Here, the watermark pattern is a binary image composed of blocks having a predetermined number of pixels far smaller than the number of pixels of the bitmap data, and is represented by turning on / off the pixels. Hereinafter, this watermark pattern is referred to as a “feature amount pattern” in order to distinguish it from other watermark patterns described later.

  FIG. 3 is a diagram showing an example of the feature value pattern, and is a diagram in which two feature value patterns WM are arranged side by side in the main scanning direction and the sub-scanning direction. The feature amount pattern WM shown in the figure is represented by 64 pixels (8 × 8 pixels), of which 60 pixels excluding on-pixels corresponding to 2 × 2 pixels in the upper left represent feature amounts. The feature amount is converted into 60-bit data by binarization processing or the like, and becomes a pixel pattern represented by ON / OFF.

  Note that the ON pixels corresponding to 2 × 2 pixels in the upper left are pixels for specifying the position of the feature amount pattern, and these are turned on regardless of the feature amount. When feature amount patterns are repeatedly arranged as shown in FIG. 3, 2 × 2 on-pixels appear regularly, so that feature amount patterns can be easily extracted from a document. An image represented by these pixels specifying the position of the feature amount pattern is hereinafter referred to as a “symbol image”. The number of pixels and the shape of the symbol image are arbitrary, but it is desirable that the symbol image can be easily extracted from the document and does not deteriorate the readability of the document.

  Returning to the description of FIG. After generating the feature quantity pattern as described above, the control unit 101 embeds this feature quantity pattern in the bitmap data (S1108). In this operation example, the control unit 101 repeatedly arranges feature amount patterns on the entire bitmap data. Thereafter, the control unit 101 outputs the bitmap data in which the feature amount pattern is embedded to the image forming apparatus 30 via the printer interface unit 105 (S1109), and ends the process of generating image data with a watermark image.

  Next, the above process will be illustrated and described. FIG. 4 is a diagram showing a case where the above-described processing is performed on a part of certain bitmap data. Here, FIG. 4A shows a part of the bitmap data representing the input document. When such bitmap data is divided into blocks of 4 × 4 pixels, for example, it is divided into 16 blocks. When ON / OFF determination is performed for each of these blocks, the result is as shown in FIG. In the drawing, a block in which the number of ON-state pixels is more than half is determined to be “ON state (1)”, and the other blocks are determined to be “OFF state (0)”. Using the 0/1 pattern determined in this way, the control unit 101 calculates a feature amount.

  Here, when the feature amount pattern based on the calculated feature amount is the feature amount pattern WM shown in FIG. 3, the bitmap data in which the feature amount pattern is embedded is as shown in FIG. As can be seen from the figure, the bitmap data output to the image forming apparatus 30 changes from the off state to the on state when compared with the bitmap data representing the input document (FIG. 4A). Although there are pixels that change, there are no pixels that change from the on state to the off state. This is because the process of embedding the feature quantity pattern is a process for calculating the logical sum of the original bitmap data and the feature quantity pattern. That is, in the process of embedding the feature amount pattern, when the feature amount pattern is repeatedly arranged in the original bitmap data, the pixel in which at least one of the original bitmap data and the feature amount pattern is on is turned on. It can be said that this is a process of outputting a set of pixels that turn off other pixels.

  Next, a description will be given of processing for detecting the presence or absence of falsification of a manuscript when a copy made based on the bitmap data output in this way is used as the manuscript. FIG. 5 is a flowchart showing processing in the first mode in which the image processing apparatus 10 of the MFP MFP detects whether the document has been tampered with. Describing along the drawing, first, the control unit 101 of the image processing apparatus 10 causes the image reading apparatus 20 to read a document in accordance with an instruction from the operator to obtain bitmap data representing the document (S1201). An attempt is made to extract a feature amount pattern from the acquired bitmap data (S1202). At this time, the control unit 101 extracts a feature amount pattern by specifying the above-described symbol image.

Note that the feature amount pattern extracted by the control unit 101 has a part of the feature amount pattern different from the original shape because, for example, other images such as characters overlap or noise is mixed. Extracted. Therefore, for example, the control unit 101 determines that the most extracted shape is a feature amount pattern, or obtains the pixel state of each block after dividing the bitmap data in units of the blocks described above, and exceeds a predetermined frequency. The shape of the feature amount pattern may be specified from the pixels that are turned on.

  Subsequently, the control unit 101 determines whether or not a watermark pattern is included in the document (S1203). This determination is made based on the number of feature amount patterns extracted in step S1202, for example. In this case, if the number of extracted feature amount patterns is equal to or greater than a predetermined value, the control unit 101 determines that the watermark pattern is included in the document. If it is determined that the watermark pattern is not included in the document, the control unit 101 causes the image forming apparatus 30 to form an image corresponding to the bitmap data (S1210).

  On the other hand, if it is determined that the watermark pattern is included in the document, the control unit 101 calculates a feature amount from the extracted feature amount pattern (S1204). This process is performed in the reverse order of the process of step S1107 for generating a feature amount pattern. Then, the control unit 101 removes the watermark pattern from the bitmap data (S1205). The process performed by the control unit 101 at this time is a process of replacing pixels corresponding to the ON pixels of the watermark pattern with the OFF pixels from the entire bitmap data.

  When the watermark pattern is removed in this way, the pixels that were in the on state before embedding the watermark pattern are also changed to the off state, and a part of the original image is lost. Therefore, the control unit 101 estimates the original image that has been lost as a result of removing the watermark pattern, and executes a process of correcting the bitmap data (S1206). Then, the control unit 101 calculates a feature amount from the bitmap data subjected to the above-described correction processing (S1207). The calculation of the feature amount is performed in the same manner as in step S1106 described above.

  The specific contents of this correction processing will be illustrated and described. FIG. 6 is a diagram for explaining this correction processing using FIG. 4 as an example. For example, when the bitmap data shown in FIG. 6A is generated, if the feature amount pattern WM shown in FIG. 3 is removed, the state shown in FIG. 6B is obtained. At this time, for each pixel in the off state, it is determined whether or not the condition that “all four pixels on the top, bottom, left, and right of the own pixel were on before the watermark pattern was removed” is satisfied. A pixel satisfying this condition is considered to be in an on state before embedding a watermark pattern. The control unit 101 corrects the bitmap data by converting an off pixel that satisfies such a condition into an on pixel. Then, the corrected bitmap data is as shown in FIG.

  As can be seen from the comparison of the bitmap data shown in FIG. 6 (c) with the bitmap data shown in FIG. 4 (a), even if such processing is executed, it is completely different from the original document. The same bitmap data is not always obtained, and some pixels may fluctuate. However, when the feature amount is calculated from the bitmap data shown in FIG. 6C, the result agrees with FIG.

  Returning to the description of FIG. Subsequently, the control unit 101 calculates the feature amount calculated in step S1207 (that is, the feature amount calculated from the bitmap data subjected to the correction process) and the feature amount calculated in step S1204 (that is, the extracted feature amount). The feature amount calculated from the pattern) is compared, and the degree of coincidence of both is determined (S1208). If it is determined that the two match, the control unit 101 outputs the bitmap data with the watermark pattern embedded to the image forming apparatus 30 (S1210) to form an image, while both If it is determined that the two do not match, it is assumed that the document has been tampered with, and predetermined error processing is executed (S1209). Here, the error processing may be, for example, processing for displaying the fact of falsification via the input / output device 40, or processing for forming an image including a mark indicating that falsification has been performed. Alternatively, a process for stopping the image formation itself may be used.

  By executing the above processing, the MFP MFP of the present embodiment makes it difficult to falsify a document and can accurately detect whether the document has been falsified. First, according to the MFP MFP of the present embodiment, since a watermark pattern is embedded on the entire surface of a document, it is possible to easily specify erasing and rewriting of characters and the like. This not only facilitates detection of manuscript falsification, but also has the effect of psychologically suppressing fraudulent acts by malicious third parties. In addition, since the MFP MFP according to the present embodiment determines whether or not falsification has occurred based on the feature amount calculated from the document itself, even if the falsification is performed skillfully, the feature amount itself changes and detects falsification. It is possible.

  In addition, the MFP MFP of the present embodiment calculates a feature amount based on a block made up of a predetermined number of pixels. In other words, the feature amount is calculated in a state where the resolution is reduced in a pseudo manner. By doing so, the MFP MFP according to the present embodiment can increase the strength against noise because the feature amount cannot be changed by a minute change that does not substantially change the document. Yes.

[Operation example 2]
Next, the operation of the MFP MFP in the second mode described above will be described. Here, as in the case of the first mode described above, the process for generating image data with a watermark image and the process for detecting the presence or absence of falsification will be described separately. Note that the operation of the MFP MFP in the second mode includes parts common to the operation in the first mode. About such a part, the step which performs the process which is common in the 1st mode is shown, etc., and explanation is omitted suitably.

  FIG. 7 is a flowchart showing a process in which the image processing apparatus 10 of the MFP MFP generates image data with a watermark image in the second mode. Referring to the figure, first, the control unit 101 of the image processing apparatus 10 receives (that is, inputs) image data via the network interface unit 103 or the scanner interface unit 104 (S2101). Next, the control unit 101 determines whether the input image data is bitmap data or PDL data (S2102). If the image data is PDL data, the control unit 101 performs rasterization processing on the image data, and generates binary bitmap data corresponding to the PDL data (S2103).

  On the other hand, when the image data is bitmap data, the control unit 101 determines the attribute of the bitmap data and performs processing for recognizing the text area and the image area (S2104). Note that the text region is a region where an image is formed by characters or fine lines, and the image region is a region where an image is formed by a natural image such as a photograph or graphics such as an illustration. The process of recognizing and distinguishing the text area and the image area may be performed by a known method such as so-called TI separation. These regions are known to have different characteristics (attributes) in frequency components and the like.

  Subsequently, the control unit 101 performs a process of specifying a region where the watermark pattern is embedded and a region where the watermark pattern is not embedded from the entire bitmap data (S2105). The region where the control unit 101 embeds the watermark pattern is, for example, the background region or the above-described text region, and the region where the control unit 101 does not embed the watermark pattern is, for example, the above-described image region. Here, the background region is a region that does not belong to any of the above-described text region and image region, and is a region that does not substantially include an ON pixel.

  The text area, the image area, and the background area described above have different attributes. Which of these areas the watermark pattern is embedded in is arbitrary and may be determined appropriately, but it is desirable that the watermark pattern not be embedded in the image area. This is because embedding a watermark pattern in the image area not only lowers the legibility of the image, but also makes it difficult to extract the watermark pattern therefrom. Note that a watermark pattern may or may not be embedded in the text area.

  Next, the control unit 101 generates information for specifying a region in which the watermark pattern is to be embedded (S2106). This information is hereinafter referred to as “region information”. As the area information, for example, for each pixel constituting the bitmap data, “1” is assigned to the pixel included in the area where the watermark pattern is embedded, and “0” is assigned to the pixel included in the area where the watermark pattern is not embedded. A numerical value obtained by encoding the bit string represented by the 0/1 pattern by the run length method can be used. After generating the region information, the control unit 101 generates a watermark pattern using the region information in the same manner as in the case of the feature amount pattern (S2107). The watermark pattern generated at this time is referred to as “region information pattern” in order to distinguish it from the feature amount pattern.

  Subsequently, the control unit 101 divides the entire bitmap data into blocks each having a predetermined number of pixels (S2108), and determines that the state of each block is either “on state” or “off state” (S2109). ). When each block is determined, the control unit 101 calculates a feature amount of the bitmap data based on the determination result (S2110), and generates a watermark pattern, that is, a feature amount pattern using the feature amount ( S2111). These processes are the same as the processes in steps S1103 to S1106 in the first mode described above.

  When the feature amount pattern and the area information pattern are generated as described above, the control unit 101 embeds these watermark patterns in the bitmap data (S2112). In this operation example, the control unit 101 embeds the watermark pattern only in the area specified when the watermark pattern is embedded in step S2105 in the bitmap data. Specifically, the area specified when the watermark pattern is embedded in step S2105 is further divided into two areas, the area information pattern is embedded in the first area, and the feature amount pattern is embedded in the second area. Perform the process. The first area is preferably a part of a predetermined area such as the upper end of the bitmap data. Thereafter, the control unit 101 outputs the bitmap data in which the watermark pattern is embedded to the image forming apparatus 30 via the printer interface unit 105 (S2113), and ends the process of generating the image data with the watermark image.

  Next, a description will be given of processing for detecting the presence or absence of falsification of a manuscript when a copy made based on the bitmap data output in this way is used as the manuscript. FIG. 8 is a flowchart showing processing in the second mode in which the image processing apparatus 10 of the MFP MFP detects the presence or absence of document falsification. Explaining along the figure, first, the control unit 101 of the image processing apparatus 10 causes the image reading apparatus 20 to read a document in accordance with an operator's instruction to obtain bitmap data representing the document (S2201). An attempt is made to extract a watermark pattern from the acquired bitmap data (S2202). At this time, the control unit 101 extracts a watermark pattern by specifying the above-described symbol image.

  Subsequently, the control unit 101 determines whether or not a watermark pattern is included in the document (S2203). This determination is made based on, for example, the number of watermark patterns extracted in step S2202. In this case, if the number of extracted watermark patterns is equal to or greater than a predetermined value, the control unit 101 determines that the watermark pattern is included in the document. If it is determined that the watermark pattern is not included in the document, the control unit 101 causes the image forming apparatus 30 to form an image corresponding to the bitmap data (S2211).

  On the other hand, when it is determined that the watermark pattern is included in the document, the control unit 101 determines whether or not the area information pattern exists in the first area (S2204). Note that the processing when it is determined that the region information pattern does not exist, that is, the processing of steps S2205 to S2211, is the same as the processing of S1204 to S1210 in the first mode, and thus description thereof is omitted.

  If it is determined that there is a region information pattern, the control unit 101 identifies region information from the region information pattern (S2212). The area information is specified by performing the process of step S2106 described above in the reverse procedure. When the area information is specified, the control unit 101 refers to the area specified by the area information and the area where the watermark pattern is actually extracted in step S2202, and compares the degree of coincidence (S2213).

  Here, the control unit 101 determines whether or not the degree of coincidence is equal to or greater than a predetermined ratio (S2214). When the area specified by the area information and the area where the watermark pattern is actually extracted are equal to or greater than a predetermined ratio, the control unit 101 has actually extracted the area specified by the area information and the watermark pattern. Assuming that the area matches, the processing from step S2205 is performed. On the other hand, if the degree of coincidence between the two is less than a predetermined ratio, the control unit 101 regards that the document has been tampered with and executes predetermined error processing (S2215). This error process may be the same process as step S2210, or a process of forming an image by blacking out a mismatch area between the area specified by the area information and the area where the watermark pattern is actually extracted. May be.

  By executing the above processing, the MFP MFP of the present embodiment can embed a watermark image in a partial area of the document. In this way, it is possible to ensure sufficient safety without impairing the visibility of images such as photographs. In addition, when the MFP MFP according to the present embodiment performs the operation in the second mode, it is possible to determine the area where the watermark image exists and the area where the watermark image does not exist based on the area information. For example, when there is no watermark pattern in a region where there should be, it is possible to specifically specify a falsified location.

[Modification]
In the above description, the present invention has been described by exemplifying one preferred embodiment, but the present invention is not limited to the above-described embodiment, and can be implemented in various other modes. It is. In the present invention, for example, the following modifications can be applied to the above-described embodiment. These modifications can be combined as appropriate.

  In the above embodiment, for convenience of explanation, the image formed on the document is a monochrome binary image. However, the present invention can be applied to a document including a color image and a document including a halftone image. Is possible. For example, for a halftone image, binarization processing may be performed on input image data, or a pixel having a gradation value larger than a predetermined threshold may be determined to be in an on state. For example, when image data composed of color components of red, green, and blue is input, a watermark pattern is generated based on one of the color components, or a luminance component is derived from the three color components. And a watermark pattern may be generated based on the luminance component.

  In the error processing described above, it is sufficient for the control unit to output information indicating that the input document has been tampered with. In such a case, the operation related to the error processing described above, that is, the formation of an image including a tampering display or a mark indicating that the falsification has been performed is triggered by the output of the above information. That's fine.

  In the case of the MFP MFP of the above-described embodiment, there is a possibility that the watermark data cannot be read accurately when the original is read in an oblique state. Therefore, when the original is read in an oblique state, image formation may be stopped or a message prompting the user to reset the original may be displayed.

1 is a block diagram illustrating a configuration of a multifunction machine according to an embodiment of the present invention. 5 is a flowchart (first mode) illustrating a process in which the image processing apparatus of the copying machine generates image data with a watermark image. It is the figure which showed an example of the feature-value pattern. FIG. 6 is a diagram illustrating processing for embedding a feature amount pattern by an image processing apparatus of a copying machine. 5 is a flowchart (first mode) illustrating a process in which the image processing apparatus of the copying machine detects whether or not a document is falsified. It is a figure for demonstrating the process which correct | amends bitmap data. 6 is a flowchart (second mode) illustrating a process in which the image processing apparatus of the copying machine generates image data with a watermark image. 6 is a flowchart (second mode) illustrating processing in which the image processing apparatus of the copying machine detects whether or not a document has been tampered with.

Explanation of symbols

MFP: MFP, 10: image processing apparatus, 101: control unit, 102: storage unit, 103: network interface unit, 104: scanner interface unit, 105: printer interface unit, 106: user interface unit, 20: image reading apparatus 30 ... Image forming apparatus, 40 ... Input / output device

Claims (3)

Input means for inputting image data representing a document by a plurality of pixels in an on state or an off state;
Feature quantity calculating means for calculating the feature quantity of the image data from the state of each pixel of the image data input by the input means;
Feature quantity pattern generation means for generating a watermark pattern represented by a plurality of pixels smaller than the total number of pixels of the image data based on the feature quantity calculated by the feature quantity calculation means;
Area specifying means for specifying a plurality of areas having different attributes from the image data input by the input means;
In addition to representing the area specified by the area specifying means, each pixel constituting the image data is calculated based on the number of pixels included in the area where the watermark pattern is embedded and the number of pixels included in the area where the watermark pattern is not embedded. Area information generating means for generating area information that is a numerical value;
Area information pattern generation means for generating a watermark pattern represented by a state of a plurality of pixels smaller than the total number of pixels of the image data based on the area information generated by the area information generation means;
The watermark pattern generated by the feature amount pattern generation unit and the region information pattern generation on a predetermined part or all of the pixels of the region specified by the region specifying unit of the image data input by the input unit Embedding means for repeatedly embedding a watermark pattern generated by the means;
An output means for outputting the write Mareta image data embedded watermark pattern by said embedding unit, at least one of the pixels in the ON state to the ON state, the OFF state of the other pixels of the image data and the watermark pattern An image processing apparatus comprising: output means for outputting a set of pixels as watermarked image data. Acquisition means for acquiring image data representing a document;
Extracting means for extracting a watermark pattern from image data;
First determination means for determining whether or not a watermark pattern is included in the image data acquired by the acquisition means based on an extraction result by the extraction means;
An area representing area information that is information for specifying an area in which a watermark pattern is embedded in an area from which the watermark pattern is extracted when the first determination unit determines that the image data includes a watermark pattern. A second determination means for determining whether or not an information pattern exists;
When it is determined by the second determination means that the area information pattern exists, the area information is specified from the area information pattern, and the area specified by the area information and the area from which the watermark pattern is extracted A comparison means for comparing the degree of matching of
Image formation for forming an image in which a mismatch area between the area specified by the area information and the area from which the watermark pattern is extracted is filled when the degree of matching compared by the comparison means is less than a predetermined ratio Means and
An image forming apparatus comprising: A program for embedding a watermark pattern in image data represented by a plurality of pixels in an on state or an off state,
On the computer,
A first process for calculating a feature amount of the image data from the state of each pixel of the input image data;
A second process for generating a watermark pattern represented by a plurality of pixels smaller than the total number of pixels of the image data based on the feature amount calculated by the first process;
A third process for identifying a plurality of regions having different attributes from the image data;
In addition to representing the area specified by the third process, each pixel constituting the image data is calculated based on the number of pixels included in the area where the watermark pattern is embedded and the number of pixels included in the area where the watermark pattern is not embedded. A fourth process for generating area information that is a numerical value;
A fifth process for generating a watermark pattern represented by a state of a plurality of pixels smaller than the total number of pixels of the image data based on the region information generated by the fourth process;
The watermark pattern generated by the second process and the watermark pattern generated by the fifth process on a predetermined part or all of the pixels of the area specified by the third process in the image data And a sixth process of repeatedly embedding
A process of outputting the write Mareta image data embedded watermark pattern by the process of the sixth, at least one of the pixels in the ON state is turned on, off other pixel state of the image data and the watermark pattern And a seventh process for outputting a set of pixels as watermarked image data.

Images