JP5447851B2 - Authenticity discrimination method of image forming body - Google Patents

Description

  The present invention relates to a technology for detecting authenticity and determining authenticity from an image forming body produced based on original image data in which binary array data is synthesized as latent image information, This technology can be used to identify and authenticate image forming bodies including securities, certificates, banknotes, stamps, stamps, stamps, passports, identification cards, passports, membership cards, and credit cards.

  Conventionally, amplitude modulation, frequency modulation, and phase modulation have been proposed as techniques for combining information with image data. These information synthesis techniques usually synthesize information in a form that cannot be perceived superficially with digital image data, and detect the authenticity of digital image data and discover unauthorized copies by detecting the synthesized information. Is to do.

  As an example of this, as an example of combining information using amplitude modulation, there is a method of combining information by performing amplitude modulation by modulo calculation (see, for example, Patent Document 1).

  An example of information synthesis using phase modulation is based on a density pattern method (see, for example, Patent Document 2).

  Further, as an example of combining information using frequency modulation, one using Fourier transform has been proposed (for example, see Patent Document 3).

  Further, there has been proposed one that detects a binary array pattern that has been converted into a latent image from a portion of the binary array data that has been generated using the Hartley transform (for example, Patent Document 4). reference).

Japanese Patent No. 3593422 Japanese Patent Laid-Open No. 9-252397 JP 2000-148015 A Japanese Patent No. 4296314

  However, the information synthesizing method using amplitude modulation disclosed in Japanese Patent No. 3593422 is easily affected by illumination unevenness when it is applied to an image, and thus it is difficult to stably detect the synthesized information. It was.

  In addition, when the information synthesizing method using phase modulation disclosed in Japanese Patent Laid-Open No. 9-25297 is applied to an image, high-precision alignment is required when detecting information, and an image scanner or image sensor is used. When an image is converted into digital data again using a camera or the like, if the pixel value of the image data cannot be correctly restored due to uneven illumination, it may be difficult to stably detect the combined information.

  In addition, in the method of synthesizing information using frequency modulation disclosed in Japanese Patent Laid-Open No. 2000-148015, when the sequence to be processed is a real value, the Fourier transform generates a complex value sequence, so half of that is extra information. There was the disadvantage of using computer memory for this extra information.

  Further, the binary image pattern detection method using the Hartley transform disclosed in Japanese Patent No. 4296314 has the disadvantages that the calculation load is large and the processing time is long because the two-dimensional Hartley transform is performed when the image data is subjected to the Hartley transform. there were. In addition, there is a disadvantage that the area of the image becomes larger than that in the case of performing the one-dimensional Hartley transform.

  An object of the present invention is to solve the above-described problems, and generates a one-dimensional real array by performing one-dimensional Hartley transform on one-dimensional lightness data read from an image forming body having an authenticity determination target image. Then, a predetermined process is performed on the one-dimensional real number array to generate discrimination data, and the discrimination data and reference data are compared and collated to determine the authenticity of the image forming body.

  Another object of the present invention is to provide a method for determining authenticity of an image forming body for the purpose of preventing forgery and alteration of the image forming body.

  (1) The present invention provides a one-dimensional lightness data acquisition step of reading an image forming body having an authenticity determination target image and acquiring one-dimensional lightness data F (i), and a one-dimensional lightness data F (i) A one-dimensional Hartley transform step for performing conversion to generate a one-dimensional real array f (u), and obtaining an absolute value of each real number of the one-dimensional real array f (u), and a one-dimensional real array fa consisting of zero or positive real numbers a numerical value conversion step for generating (u), a determination data generation step for performing predetermined filtering processing and predetermined gradation conversion on the one-dimensional real number array fa (u), and generating determination data; An authenticity determination method for an image forming body comprising at least a true / false determination step of comparing and collating with predetermined reference data to determine authenticity.

  (2) The present invention is characterized in that, in the authenticity determination step, the image forming body is determined to be authentic when the similarity between the determination data and the reference data is equal to or higher than a predetermined similarity. This is a method for determining the authenticity of a formed body.

  (3) The present invention further includes an array rearrangement step of performing a predetermined array rearrangement process on the one-dimensional real number array f (u) to generate a one-dimensional real number array f ′ (u). A method for determining the authenticity of an image forming body, characterized in that an absolute value of each real number in a one-dimensional real number array f ′ (u) is obtained and a one-dimensional real number array fa (u) composed of zero or positive real numbers is generated. .

  (4) According to the present invention, in the array rearranging step, the one-dimensional real array f (u) is divided into two equal parts to generate a first partial array and a second partial array, and the first partial array is converted into the second partial array. Move to the position of the partial array, move the second partial array to the position of the first partial array, rearrange the first partial array and the second partial array, and obtain a one-dimensional real array f ′ (u ) Is generated, the image forming body authenticity determination method.

  Since the present invention is a true / false discrimination method using the Hartley transform, it is less susceptible to illumination unevenness and the like than the conventional method using amplitude modulation, and the true / false discrimination of the image forming body is accurately performed. Is possible.

  In addition, since the present invention is a true / false discrimination method using the Hartley transform, it is necessary to detect the reference position, correct the size, and correct the rotational deviation with high accuracy as in the conventional method using phase modulation. And there is an effect of stably detecting the synthesized information.

  In addition, since the present invention uses the Hartley transform, it has an advantage that only half of the memory is used to generate the same information as the information generated by the Fourier transform, compared to the conventional method using the Fourier transform. There is an effect of realizing downsizing and cost reduction of the apparatus.

  In addition, since the present invention uses a one-dimensional Hartley transform, a band-like pattern can be used, and the effect of realizing a smaller image size and a higher processing speed than a method using a conventional two-dimensional Hartley transform. There is.

  Further, the present invention is true based on discrimination data obtained by performing predetermined processing on a one-dimensional real array generated by reading one-dimensional brightness data from an image forming body and subjecting the one-dimensional brightness data to one-dimensional Hartley transform. Since it is a technology that makes false determinations, it is necessary to identify and authenticate image forming bodies including securities, certificates, banknotes, stamps, stamps, stamps, passports, identification cards, passports, membership cards, credit cards, etc. It is possible to use in the field.

3 is a flowchart showing a first embodiment of a method for determining authenticity of an image forming body according to the present invention. 9 is a flowchart illustrating a second embodiment of the image forming body authenticity determination method according to the present invention. The figure which shows the detail of arrangement | sequence rearrangement of the one-dimensional real number array f (u) in step ST130 of FIG. (a) The figure which shows an example of the multi-value image F (i, j) which has a grain-like pattern. (b) The figure which shows an example of the binary arrangement | sequence g (u, v) which has a left-right symmetric and vertical symmetric pattern. (c) A composite image C obtained by synthesizing a two-dimensional real array obtained by Hartley transform of the multi-valued image F (i, j) and a binary array g (u, v) and performing inverse Hartley transform on the synthesized two-dimensional real array. The figure which shows an example of (i, j). The figure which shows the image forming body 100 to which the synthetic | combination image M1 (i, j) which cut out a part of synthetic | combination image C (i, j) in strip shape was provided. The figure which shows an example of the one-dimensional lightness data F (i) for 256 pixels acquired from the 10th line from the top in the strip | belt-shaped pattern of the synthesized image M1 (i, j). The figure which shows an example of the one-dimensional real number array f (u) which carried out the one-dimensional Hartley transform of the one-dimensional lightness data F (i) in step ST120 of FIG. The figure which shows an example of the one-dimensional real number array fa (u) which consists of zero or a positive real number, which was generated by obtaining the absolute value of each real number in the one-dimensional real number array f (u) in step ST140 of FIG. The figure which shows an example of the data for discrimination | determination produced | generated by performing predetermined filtering process and predetermined gradation conversion to the one-dimensional real number array fa (u) in step ST150 of FIG. The figure which shows an example of the reference data read from the data storage means in step ST160 of FIG. In step ST120 of FIG. 2, the one-dimensional lightness data F (i) is one-dimensional Hartley transformed to generate a one-dimensional real array f (u), and then in step ST130 of FIG. 2, the one-dimensional real array f (u) The figure which shows an example of the one-dimensional real number array f '(u) produced | generated by rearranging the arrangement | sequence of (). FIG. 2 shows an example of a one-dimensional real array fa (u) consisting of zero or positive real numbers generated by numerical conversion for calculating the absolute value of each real number in the one-dimensional real number array f ′ (u) in step ST140 of FIG. Figure. The figure which shows an example of the data for discrimination | determination produced | generated by performing a predetermined filtering process and predetermined gradation conversion to the one-dimensional real number array fa (u) in step ST150 of FIG. The figure which shows an example of the reference | standard data read from the data storage means in step ST160 of FIG. The figure for demonstrating the whole structure of the authenticity determination apparatus_1 which concerns on this invention. The figure for demonstrating the whole structure of the authenticity determination apparatus_2 which concerns on this invention.

  Embodiments of an image forming body authenticity determination method according to the present invention will be described below with reference to the drawings. The embodiment of the authenticity determination method of the image forming body according to the present invention is not limited to the following examples, but within the scope of the technical matters described in the claims of the present invention. If it exists, various aspects can be adopted.

  The image forming body that is the object of authenticity determination of the present invention is a forming body in which confidential information that cannot be observed in a normal observation environment is embedded in the original image. Hartley transform, synthesized with binary image embedded as confidential information, and inverse Hartley transform. The details of the image forming body are described in Japanese Patent No. 4296314 filed earlier by the applicant of the present application, and will be omitted.

  With respect to this image forming body, it is impossible to confirm the confidential information embedded under normal observation conditions, but it is possible to confirm by performing a predetermined process. This method of confirming confidential information is described in the same manner in Japanese Patent No. 4296314 filed earlier by the applicant of the present invention described in the previous section, but is described in “Problems to be solved by the invention”. As described above, the processing when reading the confidential information is to perform the two-dimensional Hartley transform, and there is a problem that the calculation load is large and the processing time is long. In view of this, the present invention performs true / false discrimination using a one-dimensional Hartley transform for the purpose of reducing the calculation load.

  Hereinafter, the above-described authenticity determination method for an image forming body according to the present invention will be described with reference to the drawings.

(First embodiment of authenticity determination method according to the present invention)
FIG. 1 is a flowchart showing a first embodiment of an image forming body authenticity determination method according to the present invention. The authenticity determination method will be described in accordance with the flowchart of FIG.

  In step ST110 of FIG. 1, one-dimensional brightness data F (i) 110 is acquired from the image forming body 100. Here, the one-dimensional lightness data F (i) 110 is a one-dimensional array in which the number of array elements is a power of two.

  In step ST120 of FIG. 1, one-dimensional Hartley transformation is performed on the one-dimensional lightness data F (i) 110 to generate a one-dimensional real number array f (u) 120. Here, a one-dimensional real array f (u) 120 which is a one-dimensional high-speed Hartley transform output of the one-dimensional lightness data F (i) 110 (i = 0, 1,..., N−1, N is the number of data). (U = 0, 1,..., N−1, N is the number of data) is defined by Equation 1.

  In step ST140 in FIG. 1, the absolute value of each real number in the one-dimensional real number array f (u) 120 is obtained, and a one-dimensional real number array fa (u) 140 composed of zero or positive real numbers is generated.

  In step ST150 of FIG. 1, the discrimination data 150 is generated by performing a predetermined filtering process and a predetermined gradation conversion on the one-dimensional real number array fa (u) 140.

  In the present invention, moving average processing is used as filtering processing in step ST150 in order to remove high-frequency noise from the one-dimensional real number array fa (u) 140.

  Filtering processing includes moving average processing, frequency domain processing, addition averaging processing, and the like. The moving average process directly processes the obtained data array and includes processing methods such as a spatial filtering process and a median filtering process.

  Frequency domain processing is processing that performs frequency transformation on the obtained data array by performing Fourier transformation, etc., applies an appropriate filter (Gauss, Hamming, Hanning window, etc.), and then performs inverse Fourier transformation, etc. is there. In addition, the addition averaging process is a process of averaging each array element after repeated input. Unlike the smoothing process, there is a feature that does not require a difference in frequency characteristics between a signal and noise. The filtering process is achieved by using one of the processing methods described above. For example, a smoothing process and an addition averaging process can be used in combination.

  The gradation conversion is a contrast amplification process for increasing the difference between the real value of a specific array element and the real value of another array element with respect to the one-dimensional real array fa (u) 140 after the filtering process. Is to do. By performing the gradation conversion, the contrast of the one-dimensional real number array fa (u) 140 is amplified, and the real value of a specific array element of the one-dimensional real number array fa (u) 140 is emphasized. This makes it possible to perform authenticity determination stably.

  Note that if tone conversion is performed on the one-dimensional real array fa (u) 140 with much noise without performing filtering processing, noise is also emphasized. Therefore, stable authenticity determination can be performed by performing gradation conversion after performing the filtering process.

  In step ST160 of FIG. 1, the discrimination data 150 is compared and collated with the reference data 160.

  The reference data 160 is the authentic authentic data that is generated by the authentic image forming body 100 by the same generation method as the determination data 150 described above. The reference data 160 can be appropriately created. For example, a plurality of (10) authentic image forming bodies 100 are used, and the reference data 160 is generated by the same generation method as the above-described determination data 150. It is also possible to create data by generating statistical data and statistically processing it. In addition, the reference data 160 may be changed to appropriate reference data as needed by a user who performs authenticity determination after the preparation, as appropriate.

  In step ST170 of FIG. 1, when the similarity between the determination data 150 and the reference data 160 is equal to or higher than a predetermined similarity, the image forming body 100 is determined to be authentic, and the authenticity determination result 200 is output. .

  As the authenticity determination method in step ST170 of FIG. 1 and FIG. 2 to be described later, for example, the similarity between the determination data 150 and the reference data 160 is quantified by using a pattern matching method such as normalized correlation, and the similarity. When the degree is equal to or higher than a predetermined similarity, the image forming body 100 is determined to be authentic.

  The degree of similarity can be set as appropriate. For example, the degree of similarity R expressed by Equation 2 can be used as an index indicating the degree of similarity.

  In Equation 2, (2) is a vector of the discrimination data 150, and (3) is a vector of the reference data 160. Further, (4) is the norm of the vector of the discrimination data 150 shown in (2), and (5) is the norm of the vector of the reference data 160 shown in (3). The norm is the square root of the sum of the squares of the vector components. The similarity R takes a value from −1.0 to +1.0, and the closer to +1.0, the higher the similarity is determined.

  Step ST170 in FIG. 1 determines that the image forming body 100 is authentic when the calculated similarity R is equal to or greater than a predetermined value, and outputs a true / false determination result 200. In step ST170, when the calculated similarity R is less than a predetermined value, the image forming body 100 is determined to be fake, and the authenticity determination result 200 is output.

(Second embodiment of authenticity determination method according to the present invention)
FIG. 2 is a flowchart showing a second embodiment of the image forming body authenticity determination method according to the present invention. The authenticity determination method will be described in accordance with the flowchart of FIG.

  In step ST110 of FIG. 2, one-dimensional lightness data F (i) 110 is acquired from the image forming body 100. Here, the one-dimensional lightness data F (i) 110 is a one-dimensional array in which the number of array elements is a power of two.

  In step ST120 of FIG. 2, one-dimensional Hartley transformation is performed on the one-dimensional lightness data F (i) 110 to generate a one-dimensional real number array f (u) 120. Here, a one-dimensional real array f (u) 120 which is a one-dimensional high-speed Hartley transform output of the one-dimensional lightness data F (i) 110 (i = 0, 1,..., N−1, N is the number of data). (U = 0, 1,..., N−1, N is the number of data) is defined by the above-described formula 1.

  In step ST130 of FIG. 2, a predetermined array rearrangement process is performed on the one-dimensional real number array f (u) 120 to generate a one-dimensional real number array f '(u) 130. The predetermined array rearrangement process will be described with reference to FIG. FIG. 3A is a diagram showing an example of the one-dimensional real number array f (u) 120 generated by using the mathematical formula 1 in Step ST120 of FIG. 2 described above. The position of the array element on the horizontal axis represents the frequency. As shown in FIG. 3A, for the one-dimensional real number array f (u) 120 before the array rearrangement process, the central portion (V) is a high frequency component and the peripheral portion (W) is a low frequency component.

  First, the one-dimensional real array f (u) shown in FIG. 3A is divided into two equal parts at the center (P) shown in FIG. 3B, and the first partial array (P1) and the second array A partial array (P2) is generated. Next, the first partial array (P1) is moved to the position of the second partial array (P2) in FIG. 3B, and the second partial array (P2) is moved to the second partial array (P2) in FIG. Move to the position of one partial sequence (P1). FIG. 3C shows the one-dimensional real number array f ′ (u) 130 after the array rearrangement process. In the one-dimensional real number array f '(u) 130 shown in FIG. 3C, the central portion (V) is a low frequency component and the peripheral portion (W) is a high frequency component.

  As a reason for rearranging the array, standard display or optical display is used when displaying a two-dimensional real array generated by Hartley transform of an image. In the standard display, display is performed such that the center is a high frequency component and the peripheral portion is a low frequency component. On the other hand, in the optical display, display is performed so that the center portion has a low frequency component and the peripheral portion has a high frequency component. The human eye is sensitive to the low frequency components of the image and easily visible, and has a characteristic that the high frequency is insensitive and difficult to see. For this reason, an optical display in which a portion where the central portion is easily visible has a low frequency component is generally used. By rearranging the array, as shown in FIG. 3C, in the one-dimensional real array f ′ (u) 130, the center (V) is a low frequency component and the periphery (W) is a high frequency component. The arrangement is suitable for optical display.

  It should be noted that the effect of the present invention is not lost unless the arrangement is rearranged, and the discrimination data 150 can be obtained without rearranging the arrangement.

  In step ST140 of FIG. 2, the absolute value of each real number in the one-dimensional real number array f '(u) 130 is obtained, and a one-dimensional real number array fa (u) 140 composed of zero or positive real numbers is generated.

  In step ST150 of FIG. 2, the above-described predetermined filtering process and predetermined gradation conversion are performed on the one-dimensional real number array fa (u) 140 to generate the determination data 150.

  In step ST160 of FIG. 2, the discrimination data 150 is compared and collated with the reference data 160.

  In step ST170 of FIG. 2, when the similarity between the determination data 150 and the reference data 160 is equal to or higher than a predetermined similarity, the image forming body 100 is determined to be authentic, and the authenticity determination result 200 is output. .

  Here, as the authenticity determination method in step ST170 of FIGS. 1 and 2, for example, using a pattern matching method such as normalized correlation, the similarity between the determination data 150 and the reference data 160 is quantified, and this When the similarity is equal to or higher than a predetermined similarity, the image forming body 100 is determined to be authentic.

  FIG. 4A shows an example of a multi-valued image F (i, j) having a grain pattern. In the example of FIG. 4A, the size of F (i, j) is 256 pixels × 256 pixels. Here, the size of F (i, j) needs to be a power of 2 × power of 2, but the shape may be square or rectangular.

  FIG. 4B shows an example of a binary array g (u, v) having a bilaterally symmetric and vertically symmetric pattern. 4 (b) has the same size as F (i, j) in FIG. 4 (a).

  4 (c) shows a two-dimensional real array f (u, v) obtained by Hartley transform of F (i, j) in FIG. 4 (a) by the mathematical formula 3 and g (u, v) in FIG. 4 (b). ) And the mathematical expression 4 and an example of a composite image C (i, j) obtained by performing inverse Hartley transform on the two-dimensional real array e (u, v) after the synthesis using the mathematical expression 5. C (i, j) in FIG. 4 (c) has the same size as F (i, j) in FIG. 4 (a). Here, in the mathematical expression of Expression 4, α is a real number.

  FIG. 5 is a diagram showing the image forming body 100 to which the composite image M1 (i, j) is given. In FIG. 5, the image forming body is a gift certificate. The composite image M1 (i, j) shown in FIG. 5 is an image obtained by cutting out a part of the composite image C (i, j) of FIG. 4C in a band shape. This is a strip-shaped image obtained by cutting out up to 20 lines. The image forming body 100 includes a composite image M1 (i, j) that is a band-shaped pattern, and acquires one-dimensional lightness data F (i) from the band-shaped pattern. In FIG. 5, the size of the composite image M1 (i, j) is 256 pixels × 20 pixels. However, the synthesized image M1 (i, j) may be cut out from any row of the synthesized image C (i, j), and the number of rows to be cut out is large enough to obtain the one-dimensional brightness data F (i). If there is, it is not limited to 20 lines and is arbitrary. In this way, the composite image M1 (i, j) in the present invention is compared with the conventional method of performing authenticity determination after obtaining the two-dimensional lightness data, for example, an image forming body 100 like a belt-like pattern. It is possible to reduce the area of the composite image M1 (i, j) to be given to.

  FIG. 6 shows an example of one-dimensional lightness data F (i) for 256 pixels acquired from the 10th row from the top in the strip pattern of the composite image M1 (i, j) in FIG.

  FIG. 7 shows an example of a one-dimensional real array f (u) obtained by performing one-dimensional Hartley transform on the one-dimensional lightness data F (i) of FIG. 6 using the mathematical formula 1 in step ST120 of FIG. As described above, the position of the array element on the horizontal axis represents the frequency, but the peripheral portion is a low frequency component and the center portion is a high frequency component.

  FIG. 8 shows a one-dimensional real array fa (zero or positive real number fa () generated by numerical conversion for calculating the absolute value of each real number of the one-dimensional real number array f (u) in FIG. 7 in step ST140 of FIG. An example of u) is shown.

  FIG. 9 shows an example of discrimination data generated by performing predetermined filtering processing and predetermined gradation conversion on the one-dimensional real array fa (u) of FIG. 8 in step ST150 of FIG. Here, with respect to the filtering process, a moving average process is performed in which the average of five neighboring array elements is obtained and smoothed in a one-dimensional real array fa (u) having 256 array elements. As for tone conversion, a process of raising the logarithm of each array element after moving average processing to the fourth power using a lookup table.

  FIG. 10 shows an example of the reference data read from the data storage means in step ST160 of FIG. In step ST170 in FIG. 1, when the similarity between the determination data in FIG. 9 and the reference data in FIG. 10 is equal to or higher than a predetermined similarity, the image forming body 100 in FIG. 1 is determined to be authentic.

  In FIG. 11, in step ST120 of FIG. 2, the one-dimensional lightness data F (i) of FIG. An example of the one-dimensional real number array f ′ (u) generated by rearranging the array of f (u) based on the above-described array rearrangement process in step ST130 of FIG. 2 is shown. In FIG. 11, the position of the array element on the horizontal axis represents the frequency, but the peripheral portion is a high-frequency component and the center portion is a low-frequency component.

  FIG. 12 shows a one-dimensional real array fa consisting of zero or positive real numbers generated by numerical conversion for calculating the absolute value of each real number of the one-dimensional real number array f ′ (u) in FIG. 11 in step ST140 of FIG. An example of (u) is shown.

  FIG. 13 shows an example of discrimination data generated by performing predetermined filtering processing and predetermined gradation conversion on the one-dimensional real array fa (u) of FIG. 12 in step ST150 of FIG. Here, with respect to the filtering process, as an example, a moving average process for obtaining and smoothing the average of five neighboring array elements in a one-dimensional real array fa (u) having 256 array elements was performed. Regarding gradation conversion, as an example, a process of raising the logarithm of each array element after moving average processing to the fourth power is performed using a lookup table.

  FIG. 14 shows an example of the reference data read from the data storage means in step ST160 of FIG. In step ST170 in FIG. 2, when the similarity between the determination data in FIG. 13 and the reference data in FIG. 14 is equal to or higher than a predetermined similarity, the image forming body 100 in FIG. 2 is determined to be authentic.

(First application example of the present invention)
Next, a first application example of the present invention will be described. FIG. 15 is a block diagram showing a configuration of the authenticity determination apparatus_1 (10000) to which the first embodiment of the authenticity determination method according to the present invention is applied. The authenticity determination apparatus_1 (10000) shown in FIG. 15 includes a one-dimensional brightness data acquisition unit 2110, a one-dimensional Hartley conversion unit 2120, a numerical value conversion unit 2140, a determination data generation unit 2150, and a storage unit 2160. The authenticity determination unit 2170 is configured.

  The one-dimensional lightness data acquisition unit 2110 in FIG. 15 includes an image scanner, an image sensor camera, and the like. Generate. Here, the one-dimensional lightness data F (i) 110 is a one-dimensional array of pixels in which the number of array elements is a power of two.

  The one-dimensional Hartley conversion unit 2120 in FIG. 15 performs a one-dimensional Hartley conversion on the one-dimensional lightness data F (i) 110 to generate a one-dimensional real array f (u) 120.

  15 obtains the absolute value of each real number in the one-dimensional real number array f (u) 120, and generates a one-dimensional real number array fa (u) 140 consisting of zero or positive real numbers.

  The discrimination data generation unit 2150 in FIG. 15 generates discrimination data 150 by performing predetermined filtering processing and predetermined gradation conversion on the one-dimensional real number array fa (u) 140.

  The storage unit 2160 in FIG. 15 includes data storage means, and writes, stores, and reads reference data 160 used for authenticity determination.

  The data storage means is means for writing and storing the reference data 160. The data storage means includes storage means such as RAM, ROM, HDD, etc. of the computer, data storage means that can be stored after writing data such as magneto-optical disk, CD-R, and the like.

  The true / false discriminating unit 2170 in FIG. 15 compares and collates the discriminating data 150 with the predetermined reference data 160 to perform authenticity discrimination, and the similarity between the discriminating data 150 and the reference data 160 is predetermined. When the degree of similarity is equal to or higher, the image forming body 100 is determined to be authentic, and the authenticity determination result 200 is output.

  In the storage unit 2160 described above, after reading the reference data 160 stored in advance in the data storage means, comparison and comparison with the determination data 150 was performed. Instead of providing data storage means in _1 (10000), the reference data 160 may be generated whenever necessary, and the prepared reference data and the determination data 150 may be compared and verified.

(Second application example of the present invention)
Next, a second application example of the present invention will be described. FIG. 16 is a block diagram showing a configuration of the authenticity determination apparatus_2 (11000) to which the second embodiment of the authenticity determination method according to the present invention is applied. The authenticity determination apparatus_2 (11000) shown in FIG. 16 includes a one-dimensional brightness data acquisition unit 2110, a one-dimensional Hartley conversion unit 2120, an array rearrangement unit 2130, a numerical value conversion unit 2140, and a determination data generation unit 2150. And a storage unit 2160 and a true / false determination unit 2170.

  The one-dimensional lightness data acquisition unit 2110 shown in FIG. 16 includes an image scanner, an image sensor camera, and the like. Generate. Here, the one-dimensional lightness data F (i) 110 is a one-dimensional array of pixels in which the number of array elements is a power of two.

  The one-dimensional Hartley conversion unit 2120 in FIG. 16 performs a one-dimensional Hartley conversion on the one-dimensional lightness data F (i) 110 to generate a one-dimensional real number array f (u) 120.

  The array rearrangement unit 2130 in FIG. 16 bisects the one-dimensional real number array f (u) 120 to generate a first partial array and a second partial array, and converts the first partial array into the second partial array. , The second partial array is moved to the position of the first partial array, the first partial array and the second partial array are rearranged, and the one-dimensional real array f ′ (u) 130 Is generated.

  16 obtains the absolute value of each real number in the one-dimensional real number array f ′ (u) 130 and generates a one-dimensional real number array fa (u) 140 composed of zero or positive real numbers.

  The discrimination data generation unit 2150 in FIG. 16 generates the discrimination data 150 by performing predetermined filtering processing and predetermined gradation conversion on the one-dimensional real number array fa (u) 140.

  The storage unit 2160 in FIG. 16 includes a data storage device, and writes, stores, and reads reference data 160 used for authenticity determination.

  The true / false discriminating unit 2170 in FIG. 16 compares and collates the discriminating data 150 with the predetermined reference data 160 to perform authenticity discrimination, and the similarity between the discriminating data 150 and the reference data 160 is predetermined. When the degree of similarity is equal to or higher, the image forming body 100 is determined to be authentic, and the authenticity determination result 200 is output.

  As mentioned above, although the embodiment of the authenticity determination method of the image forming body according to the present invention has been described based on the specific examples, the embodiment of the authenticity determination method of the image forming body according to the present invention has been described above. However, the present invention is not limited to these embodiments, and various embodiments can be adopted as long as they are within the scope of the technical matters described in the claims of the present invention.

100 Image forming body
110 One-dimensional brightness data F (i)
120 One-dimensional real array f (u)
130 One-dimensional real array f '(u)
140 One-dimensional real array fa (u)
150 Data for discrimination
160 Reference data
200 True / false discrimination result
2110 One-dimensional brightness data acquisition unit
2120 One-dimensional Hartley converter
2130 Array reordering part
2140 Value converter
2150 Data generator for discrimination
2160 Storage unit
2170 Authenticity discriminator
10000 True / False Discriminator_1
11000 Authenticity discrimination device_2
ST110 Obtaining one-dimensional brightness data
ST120 One-dimensional Hartley transform
ST130 Array reordering
ST140 Numeric conversion
Generation of ST150 discrimination data
Reading ST160 reference data
ST170 Authenticity discrimination

Claims (4)

The original image is Hartley transformed, combined with a binary image that is embedded as confidential information, and the inverse Hartley converted confidential information is a method for determining the authenticity of an image forming body that is embedded in an unobservable state under a normal observation environment. ,
A one-dimensional brightness data acquisition step of reading the image forming body and acquiring one-dimensional brightness data F (i);
A one-dimensional Hartley conversion step of performing one-dimensional Hartley conversion on the one-dimensional lightness data F (i) to generate a one-dimensional real number array f (u);
A numerical conversion step of obtaining an absolute value of each real number of the one-dimensional real number array f (u) and generating a one-dimensional real number array fa (u) composed of zero or positive real numbers;
A determination data generation step of performing predetermined filtering processing and predetermined gradation conversion on the one-dimensional real number array fa (u) to generate determination data;
A method of determining authenticity of an image forming body, comprising at least a true / false determination step of comparing and comparing the determination data with predetermined reference data to determine authenticity. In the authenticity determination step,
2. The authenticity of the image forming body according to claim 1, wherein the image forming body is determined to be authentic when the similarity between the determination data and the reference data is equal to or higher than a predetermined similarity. How to determine. A rearrangement step of performing a predetermined rearrangement process on the one-dimensional real number array f (u) to generate a one-dimensional real number array f ′ (u);
In the numerical value conversion step,
2. The image forming body according to claim 1, wherein an absolute value of each real number of the one-dimensional real number array f ′ (u) is obtained and the one-dimensional real number array fa (u) composed of zero or positive real numbers is generated. Authenticity discrimination method. In the rearrangement step,
Bisecting the one-dimensional real number array f (u) to generate a first partial array and a second partial array, moving the first partial array to a position of the second partial array, and The second partial array is moved to the position of the first partial array, and the first partial array and the second partial array are rearranged to generate the one-dimensional real array f ′ (u). The method of determining authenticity of an image forming body according to claim 3.

Images