JP4984957B2 - Image processing apparatus, image processing method, and program - Google Patents

Description

  The present invention relates to a technique for embedding data as digital watermark in image data to be printed out. In particular, the present invention can correctly detect data embedded as a digital watermark even when the printed image has expansion or contraction or inclination. The present invention relates to a possible image processing apparatus, an image processing method, and a program thereof.

  In recent years, techniques for embedding data as digital watermarks in image data have been developed. As a technology related to this, there is an invention disclosed in Patent Document 1 below.

In the watermark information embedding method disclosed in Patent Document 1, the dot pattern is configured to include at least first, second, and third dots (start point dots, horizontal reference dots, modulation dots). A value is set in the dot pattern according to the eigenvalue determined by the relative positional relationship between the second and third dots. The eigenvalue is an inner product of a vector starting from the third dot and ending with the first dot and a vector starting from the third dot and ending with the second dot.
JP 2004-128845 A

  If you print out the image data by embedding it as a digital watermark and scan the printout with a scanner to detect the embedded data, if the printout is stretched or tilted, the embedded data There is a problem that cannot be detected correctly.

  Further, in the information embedding method disclosed in Patent Document 1 described above, a signal can be restored even when the input image is distorted. However, when embedding watermark information, one data signal is assigned to a certain area. It is necessary to create and embed a unit pattern that is repeatedly embedded, and the amount of embedding is limited. In order to increase the amount of embedding, if embedding is not repeated in a certain area, the printout cannot be detected if there is expansion or contraction or inclination. Further, when the printout is scanned to detect the embedded data, it is necessary to detect and determine the unit pattern, which complicates the processing. Also, hardware cannot be easily implemented.

  The present invention has been made to solve the above problems, and an object of the present invention is to provide an image processing apparatus, an image processing method, and a program thereof capable of accurately detecting data embedded as a digital watermark. It is to be.

  In order to solve the above problems, the present invention has the following features.

In order to achieve the above object, according to one aspect of the present invention, an image processing apparatus detects data embedded in an image, the embedded data or a reference for retrieving data from the image. by detecting the standard pattern representing a pattern, and detecting means for replacing the pixels in a predetermined position of the reference pattern in said detected, respectively, in a predetermined data pixel value or a reference pixel value defined in accordance with the standard pattern, As the search range information for specifying the search range of the data pixel value, the detection means is replaced with a storage means for storing a range from a pixel at a position relatively defined with reference to the position of the reference pixel value. and with reference to the position of the reference pixel value, it searches the range or found before Symbol data pixel value defined by the Sakuhan range information, the data Characterized in that it comprises a, and the retrieval means for retrieving data from the pixel value.

Preferably, the reference pattern, it is disposed at equal intervals on the image.

Preferably, the search range information defines a range from each of a predetermined number of pixels relatively defined with reference to the position of the reference pixel value, and the extraction means includes the position of the reference pixel value. The predetermined number of data is extracted from the data pixel values obtained from each of the predetermined number of ranges defined by the search range information with reference to the respective positions defined by the search range information. It is characterized by that .

Preferably, the image processing apparatus further includes a reduction unit that performs a reduction process for narrowing a position range in which the reference pixel value replaced by the detection unit exists, and the extraction unit is subjected to the reduction process by the reduction unit. The data pixel value is searched from the range defined by the search range information with the position of the reference pixel value as a reference, and data is extracted from the data pixel value.

Preferably, the reduction unit performs a reduction process for narrowing a position range where the data pixel value replaced by the detection unit exists, and the extraction unit uses the search range information based on the position of the reference pixel value. The data pixel value after being reduced by the reduction means is searched from within the prescribed range, and data is extracted from the data pixel value.
More preferably, when the pixel value to be reduced exists in all of the specified range, the reduction means sets the specified pixel in the specified range as the pixel value to be reduced, and the reduction target pixel in the specified range. When there is even one pixel value other than the value, the specified pixel is set as a pixel value other than the pixel value to be reduced.

According to another aspect of the present invention, an image processing method is an image processing method for detecting data embedded in an image, and detects a standard pattern representing embedded data or a reference pattern for data retrieval from the image. to the pixels in a predetermined position of the reference pattern in said detected reference respectively, a detection step of replacing the predetermined data pixel value or a reference pixel value defined in accordance with the standard pattern, the position of the reference pixel values the range from the position of the pixel that is relatively defined using the search scope information for defining a range to search for data pixel values, with reference to the position of the reference pixel value is replaced in the detection step as the search this including searching the range or found before Symbol data pixel value defined by the range information, and the retrieval step of retrieving data from the data pixel values, the The features.

Preferably, the image processing method further includes a reduction process for performing a reduction process for narrowing a position range where the reference pixel value replaced in the detection process exists, and the extraction process is executed after the reduction process is performed in the reduction process. It is the fact that to said.

More preferably, in the reduction step, a reduction process for narrowing a position range where the data pixel value replaced in the detection step exists is performed, and the extraction step is executed after the reduction process in the reduction step. the shall be the feature.

According to still another aspect of the present invention, the program is a program that causes a computer to detect data embedded in an image, and the computer uses reference pixels as search range information that defines a range for searching for data pixel values. A range from a pixel at a position relatively defined with respect to the position of the value is stored, and a standard pattern representing a reference pattern for embedded data or a data extraction position is detected from the image, and the detection is performed. the pixels in a predetermined position of the reference pattern in the respective, a detection step of replacing the predetermined data pixel value or a reference pixel value defined in accordance with the standard pattern, position of the reference pixel value is replaced in the detection step the basis, searches the range or found before Symbol data pixel value defined by the search range information, the data Wherein the executing and the retrieval step of retrieving data from the pixel value, to the computer.

Preferably, the program further causes the computer to execute a reduction process for performing a reduction process for narrowing a position range in which the reference pixel value replaced in the detection process exists before the extraction process is executed. The

Preferably, the program further causes the computer to execute a reduction process for performing a reduction process for narrowing a position range where the data pixel value replaced in the detection process exists before the extraction process is executed. The
More preferably, in the reduction step, when the pixel value to be reduced exists in all of the specified range, the program sets the specified pixel in the specified range as the pixel value to be reduced, and reduces the reduced value within the specified range. When at least one pixel value other than the target pixel value exists, the computer is caused to execute a process of setting the specified pixel to a pixel value other than the reduction target pixel value.

  According to the present invention, the embedded data pattern and the reference pattern are detected and replaced with a predetermined data pixel value or a reference pixel value, respectively, and then from within a predetermined range determined based on the position of the reference pixel value. The data pixel value is searched, and data is extracted from the data pixel value.

  This makes it possible to search each data pixel value from within a predetermined range based on the position of the reference pixel value, even when the printed image has an expansion / contraction or inclination, and therefore data embedded as a digital watermark. Can be detected correctly.

(First embodiment)
FIG. 1 is a diagram illustrating an example of a system configuration of an image processing apparatus according to the first embodiment of the present invention. A schematic configuration of the image processing apparatus according to the present embodiment will be described with reference to FIG.

(Schematic configuration of image processing apparatus)
FIG. 1 shows a case where an image processing apparatus is configured using a personal computer (hereinafter referred to as a PC), and includes a mouse 11, a keyboard 12, a monitor 13, an external storage device 14, a scanner 15, PC16.

  The PC 16 has the same configuration as general computer hardware, and a function of an image processing apparatus to be described later is realized by executing an image processing program 17 by a CPU (Central Processing Unit) (not shown). .

  The image processing program 17 is generally used by installing on a PC 16 what is recorded on an external recording medium such as a CD or DVD. The installed image processing program is stored in a memory or the like inside the PC 16.

  The mouse 11 and the keyboard 12 are used as input devices, respectively, and are used when the user gives various instructions when starting the image processing program 17 and executing the image processing program 17.

  The monitor 13 is used to display image data read by the scanner 15, and the user proceeds with image processing by giving an instruction to the image processing program 17 while referring to the contents displayed on the monitor 13. .

  The external storage device 14 is used to store image data read by the scanner 15. The image processing program 17 may be stored in the external storage device 14, and the PC 16 may load the image processing program 17 from the external storage device 14 into an internal RAM (Random Access Memory) and execute it.

  The scanner 15 reads the image data in which the data is embedded, and outputs the read image data to the PC 16. The PC 16 performs image processing by performing processing to be described later on the image data received from the scanner 15.

  FIG. 2 is a diagram for functionally explaining the image processing apparatus shown in FIG. The PC 16 includes an input / output interface 24, a CPU / memory 25, and a storage device 26. In addition, the storage device 26 includes an OS (Operating System) 27 and an image processing unit 28 realized by the image processing program 17. The image processing program 17 operates on the OS 27 and realizes the function of the image processing unit 28.

  The image processing unit 28 controls the keyboard 11, mouse 12, monitor 13, and scanner 15 by inputting and outputting data via the OS 27 and the input / output interface 24. The image processing unit 28 receives a user instruction 21 from the keyboard 11 or the mouse 12, receives a scanned image 23 from the scanner 15, and displays an image data display 22 on the monitor 13 to perform image processing to be described later.

  FIG. 3 is a diagram showing another example of the system configuration of the image processing apparatus according to the first embodiment of the present invention. FIG. 3 shows a case where an image processing apparatus is configured using an MFP (Multi Function Peripheral), which includes an operation panel unit 31, a scanner unit 32, a printer unit 33, and an MFP main body 34.

  The MFP main body 34 includes an image processing circuit 35 and the like. The image processing circuit 35 receives a user instruction from the operation panel 31 and performs image processing to be described later while controlling the scanner unit 32 and the printer unit 33.

(Schematic flow of image processing)
FIG. 4 is a flowchart for explaining an outline of processing of the image processing apparatus according to the first embodiment of the present invention. A schematic flow of an image processing method from data embedding to data detection will be described with reference to FIG.

  First, step S11 is a data embedding process step. Embedded data 42 to be embedded in the image data 41 is acquired and embedded as a dot pattern in the background of the image data 41.

  The embedded image in which the embedded data 42 is embedded as a dot pattern is printed out, and a printed matter 43 is created.

  FIG. 5 is a diagram illustrating a dot pattern when embedding data 42 is embedded. FIG. 5A shows a dot pattern as a data pattern representing bit data “0”. By arranging black dots at the positions shown in FIG. 5A among 16 × 16 pixels, data “0” is obtained. ".

  FIG. 5B shows a dot pattern as a data pattern representing the bit data “1”. By arranging black dots at positions shown in FIG. 5B among 16 × 16 pixels, data Expresses “1”.

  FIG. 5C shows a dot pattern as a reference pattern representing a reference position for data extraction. By arranging black dots at the positions shown in FIG. 5C among 16 × 16 pixels, Represents a reference position for data retrieval.

  FIG. 6 is a diagram illustrating an example of embedded data. As shown in FIG. 6, 24-bit data and a data extraction reference pattern are arranged in an 80 × 80 pixel area. The data extraction reference pattern is located at the center of the 24-bit data.

  FIG. 7 is a diagram showing a dot pattern when the embedded data shown in FIG. 6 is actually printed out. As shown in FIG. 7, a dot pattern as a data extraction reference pattern is arranged in a central 16 × 16 pixel region of 80 × 80 pixels, and the other regions correspond to the embedded data shown in FIG. A data pattern, that is, a dot pattern indicating data “0” or data “1” is arranged.

  FIG. 8 is a diagram illustrating an example of a printout (printed material 43) when printing is performed with data embedded in the background of image data. The dot pattern shown in FIG. 7 is repeatedly arranged on one surface of the paper, and is combined with image data such as characters to be actually printed out and printed out. When this dot pattern is embedded, for example, at 600 dpi, this data cell is a very small dot with a side of 0.068 cm (16/600 inches). Therefore, it seems to the user that a gray image 61 has been added to the background of the printout.

  Step S12 is a step including a detection step and an extraction step as data detection processing, and details of this step will be described later with reference to FIG.

(Functional configuration of image processing)
FIG. 9 is a block diagram illustrating a functional configuration of the image processing apparatus illustrated in FIG. 1 or 3.

  This image processing apparatus is detected by an image reading unit 51 that reads an image by scanning a printed matter 43, a pattern detection unit 52 as a detection unit that performs pattern matching on the read image, and a pattern detection unit 52. A reduction processing unit 53 as a reduction unit that performs dot reduction processing on the received data, and a data extraction unit 54 as an extraction unit that extracts embedded data from the data subjected to dot reduction processing.

  The function of each processing means will be described together with the image processing procedure.

(Detailed flow of image processing)
FIG. 10 is a flowchart for explaining the detailed processing procedure of step S12 shown in FIG. The processing procedure of the image processing method for embedded data detection according to the present embodiment will be described with reference to FIG.

  First, in step S <b> 20, the image reading unit 51 scans the printed material 43 by the scanner 15 or the scanner unit 32 and acquires a scanned image 44.

  It is not always necessary to use a reading device such as the scanner 15 or the scanner unit 32 for acquiring the data embedded image, but as an external device connected via the input / output interface 24 or electronic data read from an external storage medium. It may be in the form of obtaining.

  In step S21 as a detection process, the pattern detection unit 52 scans the image data 44 obtained by the image reading unit 51 or by other means to obtain a 17 × 17 pixel pattern (hereinafter referred to as an input pattern). Extract and perform pattern matching on the input pattern.

<Detection process>
As an example of the pattern matching process, a case where data is detected using simple similarity will be described below. However, the present invention is not limited to this as long as the method can detect embedded data.

  The simple similarity is expressed by the following equation (1) when the standard pattern is c = (c1, c2,..., C289) and the input pattern is x = (x1, x2,..., X289). In this method, the angle formed by two vectors is obtained, and the smaller the smaller, the more similar the images are determined. “·” Indicates an inner product of vectors, and “||” indicates the magnitude of the vector.

c · x / (| c | × | x |) (1)
The pattern detection unit 52 calculates the cos of the angle calculated by the equation (1), and determines that the standard pattern and the input pattern are closer to “1”.

  FIG. 11 is a diagram for explaining a standard pattern. FIG. 11A shows a standard pattern of data “0”. FIG. 11B shows a standard pattern of data “1”. FIG. 11C shows a standard pattern of the data extraction reference pattern.

  If the cos value when the data “0” shown in FIG. 11A is used as the standard pattern is equal to or greater than a predetermined threshold value, the pattern detection unit 52 calculates the pixel value of the center pixel of 17 × 17 pixels. “0” (hereinafter, this pixel is referred to as a data 0 pixel as a data pixel corresponding to the data “0”). Hereinafter, it is represented by a dark gray dot 0 in the drawing.

  Further, if the cos value when the data “1” shown in FIG. 11B is used as the standard pattern is equal to or greater than a predetermined threshold value, the pattern detection unit 52 is the pixel of the center pixel of 17 × 17 pixels. The value is “1” (hereinafter, this pixel is referred to as a data pixel corresponding to the data “1”). Hereinafter, it is represented by a light gray dot 1 in the drawings.

  Further, the pattern detection unit 52 determines the center pixel of 17 × 17 pixels if the cos value when using the data extraction reference pattern shown in FIG. 11C as a standard pattern is equal to or greater than a predetermined threshold value. The pixel value is “2” (hereinafter, this pixel is referred to as a pixel corresponding to the reference pattern and is referred to as a reference pixel for data extraction). Hereinafter, it is represented by black dots 2 in the drawings.

  Further, the pattern detection unit 52 sets the pixel value of the center pixel of 17 × 17 pixels to “3” if any of the standard patterns is used and the cos value is less than the predetermined threshold value. Hereinafter, it is represented in white in the drawings.

  The pattern detection unit 52 extracts the input pattern of 17 × 17 pixels while shifting the scanned image of the printed matter 43 by one pixel in the horizontal direction and the vertical direction, and performs pattern matching to convert the image data into the pixel values described above. replace.

  FIG. 12 is a diagram schematically illustrating an example of image data after being replaced with pixel values by the pattern detection unit 52. As shown in FIG. 12, the data embedded in the image data is replaced with pixel values “0” to “3”.

  In FIG. 12, reference numeral 0 indicates a dot composed of data 0 pixels, and reference numeral 1 represents an example of a dot composed of data 1 pixels. Reference numeral 2 is a dot made up of a reference pixel, which repeats in a grid pattern with a constant period in the vertical and horizontal directions as shown in the figure. This corresponds to the fact that the 80 × 80 pixel embedding pattern shown in FIG. 6 (the central portion of which is a reference pattern and replaced with a reference pixel) is repeatedly arranged over the paper. The data 0 pixel and the data 1 pixel are also arranged according to the pattern of FIG. All other white background pixels have a pixel value of “3”.

  In FIG. 12, the dots having the respective codes from 0 to 2 are described, but each dot actually includes a plurality of pixels. For example, the dot 2 is composed of a plurality of reference pixels. This is shown in FIG.

  FIG. 13B is a partially enlarged view showing an example of the dot 2 in FIG. Each cell represents a pixel, the hatched pixel is a reference pixel, that is, a pixel having a reference pixel value of 2, and the white pixel is a pixel having a pixel value of 3 that is neither a reference pixel nor a data pixel. In the case of this figure, the dot 2 is composed of 13 reference pixels, and the position range of the reference pixel value is an area indicated by hatching.

  The expansion of the position range of the reference pixel value is generally caused by the reason described below.

  a. When the image in which the data is embedded is printed out and becomes the printed matter 43, each dot forming the reference pattern is expanded by printing and becomes a thick dot state, and the position of the reference pixel detected based on the dot is also thick. It becomes a dot.

  b. When the printed material 43 is read by a scanner or the like to obtain scanned image data 44, the distance between the dots constituting the reference pattern is locally expanded or contracted due to the floating of the printed material from the scanner surface, and the relative position. A relationship shift occurs. As a result, the range of the reference pixel position is expanded during pattern detection.

  As a result, the dot after being replaced with the reference pixel value in the pattern detection, that is, the position range of the reference pixel value is wider than the position range of the reference pixel value when there is no influence as described above. It will be reproduced as it is held.

  Note that what has been described above using the reference pixel as an example also applies to the data pixel. That is, similarly, a dot is composed of a plurality of data pixels, and the position range of each dot, that is, each data pixel value is expanded for the same reason.

  For example, in the case of the reference pixel, since the dot is thick, the reference pixel position cannot be specified, and the data extraction filter processing described later is repeated at a plurality of reference pixel positions. Will be dropped. Further, in the case of data pixels, there is a risk that dot thickening is added to the positional deviation at the time of reading, so that the data data protrudes into adjacent data cells at the time of data extraction filter processing described later, and erroneous data extraction occurs.

  In order to deal with such a point, reduction processing may be performed to narrow a position range where each pixel value exists. Details of the reduction processing will be described below.

<Reduction process>
Next, in step S <b> 22 as a reduction process, the reduction processing unit 53 performs a reduction process for narrowing the position range where the pixel value replaced by the pattern detection unit 52 exists. However, this reduction process can be omitted and the process can proceed to the next extraction step.

  FIG. 13A is a diagram illustrating an example of a filter used for the reduction process. This reduction process may be performed for both the data pixel and the data extraction reference pixel, or may be performed for either one of them.

  For example, when the reduction process is performed on the reference pixel, this reduction process filter is used for the pixel 63 indicated by other diagonal lines even if the black pixel 62 (target pixel) in FIG. 13A is the reference pixel (pixel value 2). If any of the pixels is not a reference pixel, it is determined that the target pixel 62 is not a reference pixel. That is, the pixel value of the target pixel is replaced with another pixel (pixel value 3). The same applies to the case of reducing the data pixel.

  With this reduction filter processing, for example, if the dot is a reference pixel, the pixel at the outer peripheral edge of the dot area is considered not to be the reference pixel and deviates from the dot area. It will be narrowed towards. The same applies to data pixels.

  The reduction processing unit 53 uses this reduction processing filter to perform reduction processing on all the pixels of the image data after being replaced with pixel values by the pattern detection unit 52. That is, the filter processing is performed on the entire region while shifting the region to which the reduction processing filter is applied to the image data after being replaced with the pixel value in the horizontal and vertical directions pixel by pixel.

  Note that the reduction processing filter is not limited to the 9 × 3 filter of 3 × 3 size shown in FIG. 13A, and is changed to a 25 × 5 filter of 5 × 5 size, for example, depending on the reduction amount. It is also possible.

  FIG. 14 is a diagram showing the image data after performing the reduction process. As shown in FIG. 14, the dot is reduced by the reduction process in both the data extraction reference pixel, the data 0 pixel, and the data 1 pixel. Except for being reduced, the positions of the dots indicated by the respective symbols are the same as those in FIG.

  As described above, if the position range where the reference pixel for data extraction exists is narrowed by performing the reduction process, the number of times of repeating the data extraction filter processing described later with respect to the reference pixel position is set as the reference pixel. The position range can be reduced by narrowing, and the data extraction process can be performed at high speed.

  For data pixels, if the position range where each data pixel exists is narrowed by performing a reduction process, the data extraction process described later will be performed even if the position of the dot of each data pixel is shifted. Therefore, it is possible to suppress the risk of causing an erroneous data extraction by protruding into an adjacent data extraction range.

<Removal process>
Next, step S23 to step S26 function as an extraction process. Also in the extraction process, filter processing using a data extraction filter is performed.

  In step S23, the data extraction unit 54 determines whether the pixel of interest is a data extraction reference pixel while shifting the image data in the horizontal direction and the vertical direction pixel by pixel.

  If the target pixel is the reference pixel for data extraction (step S23: Yes), step S24 is executed, and data is extracted from the periphery of the target pixel by filter processing. If the pixel of interest is not a reference pixel for data extraction (step S23: No), nothing is done (step S25) and the process proceeds to step S26.

  FIG. 15 is a diagram illustrating an example of a filter used when the data extraction unit 54 extracts data. This filter searches for data within a predetermined range 67 in the vicinity of 24 pixels (black circle 66) that are 16 pixels apart from each other with the center pixel (white circle 65) as the target pixel. is there.

  By searching within the predetermined range 67 as described above, even when the pixel interval in the read image data is slightly changed due to a slight floating of the paper at the time of image reading, an appropriate value is obtained. Data can be searched from within the range and detected correctly.

  That is, the data extraction unit 54 shifts the filter shown in FIG. 15 by one pixel in the horizontal direction and the vertical direction with respect to the image data, and when the target pixel 65 is the data extraction reference pixel, the data extraction unit 54 is located at a position indicated by a black circle. A predetermined range 67 centering on the pixels 66 at the location is searched. And data is taken out by acquiring the minimum value within the range.

  The acquisition of the minimum value as data is based on the method of giving the pixel value to the data pixel and the reference pixel. That is, in this embodiment, the data pixel value “0” or the data pixel value “1” is set for the data 0 pixel or the data 1 pixel, the reference pixel value “2” is set for the reference pixel, and the other pixels. Is given a pixel value of “3”. Therefore, the data to be extracted is 0 or 1, and if the minimum value is acquired from the possible pixel values 0 to 3, one of the data 0 or 1 that should be in the range can be acquired.

  FIG. 16 is a diagram for explaining processing when data is extracted by the filter shown in FIG. As shown in the upper left of FIG. 16, when the pixel of interest at the center of the data extraction filter 71 is a data extraction reference pixel, 24 pieces of data around it are extracted.

  Further, a reading error has occurred at a location 72 indicated by a circle in the upper left of FIG. 16 (an enlarged view is shown in FIG. 17A), and the pixel value is erroneously changed to “ 3 ”. As described above, even when there is a reading error in a part of the data, it is possible to determine that the data at the part is unknown by using the filter shown in FIG.

  Further, it is indicated that data is erroneously determined at a location 73 indicated by a circle on the lower right in FIG. 16 (an enlarged view is shown in FIG. 17B). That is, the data pixel is generated at a position where the data pixel should not exist. As described above, even when some data is erroneously determined, the data at that portion can be skipped by using the filter shown in FIG.

  17 (a) and 17 (b) are enlarged views when the pixel of interest in the center of the data extraction filter 71 corresponds to a data extraction reference pixel (indicated by reference numeral 2) that exists at a fixed period. It also shows that data extraction processing is performed.

  FIG. 18 is a diagram illustrating another example of a filter used when the data extraction unit 54 extracts data. 15 extracts 5 × 5 data, but FIG. 18 extracts 6 × 6 data. When extracting 6 × 6 data, there is no center pixel, and therefore, as shown in FIG. 18, one of the four white circles 68 at the center is used as a target pixel, and a reference pixel for data extraction is arranged. Data is extracted from other pixels.

  Returning to the description of the flowchart shown in FIG. Finally, in step S26, the data extraction unit 54 determines whether there is a next pixel to be processed. If there is a next pixel (step S26: Yes), the process returns to step S23 and the subsequent processing is repeated. If there is no next pixel (step S26: No), the process is terminated.

  As described above, according to the image processing apparatus in the present embodiment, image data is replaced with predetermined pixel values by pattern matching processing, and data pixel positions are searched from within a predetermined range by filter processing for data extraction. Since the embedded data is taken out, the data can be detected correctly even when the pixel spacing fluctuates slightly because the paper is slightly floating when scanning the image. It has become possible.

  In addition, detection processing by pattern matching, reduction processing by a reduction filter, and data extraction processing can all be performed, and processing can be performed simply by providing a line buffer corresponding to the filter size. It became possible to implement hardware on a scale.

  Even when the data extraction filter processing is realized by software, the data extraction processing only needs to be performed when the target pixel is the reference pixel for data extraction. Data can be detected at high speed.

(Second Embodiment)
The image processing apparatus according to the second embodiment of the present invention is different from the image processing apparatus according to the first embodiment in that the processing is performed when image data is scanned at an inclination. Therefore, detailed description of overlapping configurations and functions will not be repeated.

  In the present embodiment, it is assumed that the inclination of the scan image is known in advance, for example, by a method using a reference mark for inclination detection.

  FIG. 19 is a diagram illustrating an example of a standard pattern used when the scan image is tilted. FIG. 19A shows a standard pattern of data “0”. FIG. 19B shows a standard pattern of data “1”. FIG. 19C shows a standard pattern of the data extraction reference pattern.

  As shown in FIG. 19, the number of pixels of the standard pattern is 21 × 21 due to the inclination of the scan image. The pattern detection unit 52 extracts an input pattern of 21 × 21 pixels while shifting the scanned image of the printed material 43 pixel by pixel in the horizontal direction and the vertical direction, and performs pattern matching with the standard pattern shown in FIG. The image data is replaced with the above pixel values.

  A plurality of standard patterns corresponding to the inclination of the scan image are prepared in advance, and an optimum one is selected and used from the plurality of standard patterns according to the detected inclination of the scan image.

  FIG. 20 is a diagram illustrating an example of image data after being replaced with pixel values by the pattern detection unit 52 when the scan image is tilted. As shown in FIG. 20, the data embedded in the image data is tilted as a whole and replaced with pixel values “0” to “3”.

  Even when the reduction filter processing by the reduction processing unit 53 is performed, in FIG. 20, the data 0 pixel and the data 1 pixel are reduced while being entirely inclined by the filter processing.

  FIG. 21 is a diagram for explaining processing when data is extracted when the scan image is tilted. As shown in the upper left of FIG. 21, a data extraction filter 74 inclined according to the inclination of the scan image is used, and when the pixel of interest at the center of the filter is a reference pixel for data extraction, Data is extracted from a predetermined range as the center (see FIG. 22).

  FIG. 22 shows an example in which the data extraction filter shown in FIG. 15 is tilted.

(H / W image processing)
FIG. 23 is a diagram illustrating an example of a hardware configuration of filter processing. This filter corresponds to the data extraction filter shown in FIG. 22, and includes 104 line buffers 1 to 104, 105 shift register groups, and a filter processing arithmetic circuit.

  Each of the line buffers 1 to 104 holds one line of image data. This data extraction filter performs a filtering process on data of 105 × 105 pixels. Since the data on the top line is directly input to the shift register group, the line buffer corresponding to that line is used. Is not provided.

  The shift register group sequentially shifts data for 105 pixels in the horizontal direction, and data is shifted pixel by pixel in synchronization with a clock (not shown).

  The filter processing arithmetic circuit performs the data extraction process described in the extraction step in FIG. 10 on the data of 105 × 105 pixels held in the shift register group.

  As described above, since the filtering process can be performed only by providing the line buffer corresponding to the filter size as the image memory, it is possible to easily implement hardware.

  Further, the filter of the pattern detection unit 52 can be realized with the same configuration, and a pattern matching process can be performed only by providing a line buffer for 24 lines as image data in order to hold data of 25 × 25 pixels. The reduction filter of the reduction processing unit 53 can also be realized with the same configuration, and reduction processing can be performed only by providing a line buffer for two lines as image data.

  FIG. 24 is a diagram for explaining pipeline processing when the image processing apparatus according to the second embodiment of the present invention is configured by hardware. When the tilt angle of the scan image is the maximum, the filter of the pattern detection unit 52 has a size of 25 × 25 and is configured by a line buffer for 24 lines, so that a delay for 24 lines occurs.

  Therefore, a delay of 24 lines occurs until the data extraction unit 54 starts the data extraction process (however, when the reduction filter process of FIG. 13A is performed, a delay of 2 lines is added). .

  In this way, hardware corresponding to pipeline processing can be implemented by providing a delay line buffer.

  As described above, according to the image processing apparatus of the present embodiment, since the tilted filter is used corresponding to the tilt of the scan image, the filter processing can be performed at high speed.

  In addition, when the pattern detection unit 52, the reduction processing unit 53, and the data extraction unit 54 are configured by hardware, pipeline processing can be performed, so that the processing speed can be greatly improved.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

It is a figure which shows an example of the system configuration | structure of the image processing apparatus in the 1st Embodiment of this invention. It is a figure for demonstrating more functionally the image processing apparatus shown in FIG. It is a figure which shows another example of the system configuration | structure of the image processing apparatus in the 1st Embodiment of this invention. 3 is a flowchart for explaining an outline of processing of the image processing apparatus according to the first embodiment of the present invention. It is a figure which shows a dot pattern when embedding bit data. It is a figure which shows an example of embedding data. FIG. 7 is a diagram showing a dot pattern when the embedded data shown in FIG. 6 is actually printed out. It is a figure which shows an example of the printout when data is embedded and printed in the background of image data. FIG. 4 is a block diagram illustrating a functional configuration of the image processing apparatus illustrated in FIG. 1 or FIG. 3. It is a flowchart for demonstrating the detailed process sequence of step S12 shown in FIG. It is a figure for demonstrating a standard pattern. It is a figure which shows an example of the image data after being replaced by the pixel value by the pattern detection part. It is a figure which shows an example (b) of a position range of a reference pixel after replacing an example (a) of a filter used for a reduction process and a reference pattern with a reference pixel value. It is a figure which shows the image data after performing a reduction process. It is a figure which shows an example of the filter used when the data extraction part 54 extracts data. It is a figure for demonstrating the process when taking out data with the filter shown in FIG. It is a one part enlarged view of FIG. It is a figure which shows another example of the filter used when the data extraction part 54 extracts data. It is a figure which shows an example of the standard pattern used when a scan image inclines. It is a figure which shows an example of the image data after being replaced by the pixel value by the pattern detection part 52, when a scan image is inclined. It is a figure for demonstrating the process at the time of taking out data when a scan image inclines. It is a figure which shows an example of the filter for data extraction used when a scan image inclines. 3 is a diagram illustrating an example of a hardware configuration of a reduction processing unit 53. FIG. It is a figure for demonstrating the pipeline process at the time of comprising the image processing apparatus which concerns on this embodiment with hardware.

Explanation of symbols

11 mouse 12 keyboard 13 monitor 14 external storage device 15 scanner 16 PC
17 Image processing program 21 User instruction 22 Image data display 23 Scanned image 24 Input / output interface 25 CPU / memory 26 Storage device 27 OS
28 Image processing section 31 Operation panel section 32 Scanner section 33 Printer section 34 MFP main body 41 Image data 42 Embedded data 43 Printed matter 44 Scanned image data 51 Image reading section 52 Pattern detection section 53 Reduction processing section 54 Data extraction section

Claims (14)

An image processing device for detecting data embedded in an image,
A standard pattern representing embedded data or a reference pattern for data retrieval is detected from the image, and pixels at predetermined positions in the detected standard pattern are respectively determined according to the standard pattern. Detecting means for replacing the data pixel value or the reference pixel value with
Storage means for storing a range from a pixel at a position relatively defined with respect to the position of the reference pixel value as search range information for defining a range for searching for the data pixel value;
With reference to the position of the reference pixel value is replaced by the detection means, the search range information by searching the range or found before Symbol data pixel value defined, and the retrieval means for retrieving data from the data pixel value And an image processing apparatus.   The image processing apparatus according to claim 1, wherein the reference patterns are arranged on the image at equal intervals. The search range information defines a range from each of a predetermined number of pixels relatively defined with reference to the position of the reference pixel value,
Each of the positions defined by the search range information from the data pixel values obtained from each of the predetermined number of ranges defined by the search range information on the basis of the position of the reference pixel value. The image processing apparatus according to claim 1, wherein the predetermined number of pieces of data is extracted in correspondence with the data. Reduction means for performing a reduction process for narrowing a position range where the reference pixel value replaced by the detection means exists;
The extraction means searches for the data pixel value from the range defined by the search range information with reference to the position of the reference pixel value after being reduced by the reduction means, and from the data pixel value the image processing apparatus according to claim 1, characterized in that retrieving the data. The reduction means performs a reduction process for narrowing a position range where the data pixel value replaced by the detection means exists,
The extraction means searches the data pixel value after being reduced by the reduction means from within the range defined by the search range information with reference to the position of the reference pixel value, and from the data pixel value The image processing apparatus according to claim 4 , wherein data is extracted. The reduction means uses the specified pixel in the specified range as the pixel value to be reduced when the pixel value to be reduced exists in the specified range, and a pixel other than the reduced pixel value in the specified range. The image processing apparatus according to claim 4 , wherein when at least one value exists, the prescribed pixel is set to a pixel value other than the pixel value to be reduced. An image processing method for detecting data embedded in an image,
A standard pattern representing embedded data or a reference pattern for data retrieval is detected from the image, and pixels at predetermined positions in the detected standard pattern are respectively determined according to the standard pattern. A detection process for replacing the data pixel value or the reference pixel value with
Using the search range information that defines the range in which the data pixel value is searched for the range from the pixel at a position that is relatively defined with respect to the position of the reference pixel value, the reference pixel value that has been replaced in the detection step position as a reference, the searches the range or found before Symbol data pixel value defined by the search range information, the image processing method characterized by including: a take-out step of taking out the data from the data pixel values. A reduction step of performing a reduction process for narrowing a position range where the reference pixel value replaced in the detection step exists;
The image processing method according to claim 7 , wherein the extraction step is executed after the reduction process is performed in the reduction step. In the reduction process, a reduction process for narrowing a position range where the data pixel value replaced in the detection process exists is performed.
The image processing method according to claim 8 , wherein the extraction step is executed after the reduction process in the reduction step. In the reduction step, when the pixel values to be reduced exist in all the specified range, the specified pixels in the specified range are set as the pixel values to be reduced, and the pixels other than the pixel values to be reduced are in the specified range. The image processing method according to claim 8 or 9 , wherein when at least one value exists, the prescribed pixel is set to a pixel value other than the pixel value to be reduced. A program that causes a computer to detect data embedded in an image,
The computer stores a range from a pixel at a position relatively defined with reference to a position of a reference pixel value as search range information defining a range for searching for a data pixel value,
A standard pattern representing embedded data or a reference pattern for data retrieval is detected from the image, and pixels at predetermined positions in the detected standard pattern are respectively determined according to the standard pattern. A detection process for replacing the data pixel value or the reference pixel value with
With reference to the position of the reference pixel value is replaced in the detection step, the range information by searching the range or found before Symbol data pixel value defined, and the retrieval step of retrieving data from the data pixel values, a program characterized by causing the computer to perform the. Before performing the removal step,
The program according to claim 11, characterized in that to execute further the computer a reduction step of performing reduction processing of narrowing the range of positions wherein the reference pixel value is replaced in the detection step is present. Before performing the removal step,
The program according to claim 12, characterized in that to execute further the computer a reduction step of performing reduction processing of narrowing the range of positions said data pixel value is replaced in the detection step is present. In the reduction step, when the pixel value to be reduced exists in all within the specified range, the specified pixel in the specified range is set as the pixel value to be reduced, and the pixels other than the pixel value to be reduced are in the specified range. value to execute the processing of the specified pixel as a pixel value other than the pixel values of the reduced object to the computer when there is at least 1, a program according to claim 12 or 13.