Program, information processing method, and information processing system

Abstract

The present disclosure relates to a program, an information processing method, and an information processing system with which it is possible to appropriately determine, using an image obtained by imaging the face of a card or the like, whether or not the printed text or photograph is a genuine element that has been printed using a reversible thermosensitive recording medium. Texture information is extracted by applying adaptive filter processing to a captured image of the surface of a printed matter that has been imaged at a predetermined resolution or greater, subtracting the filtered image from the original image, and then normalizing the resulting image using the variation in peripheral pixels. The texture information is subjected to frequency analysis. On the basis of the frequency analysis results, the ratio of a main angle at which the frequency is detected to the sum of all angles is obtained as a score, and authenticity is determined by comparing said score against a predetermined threshold. The present invention can be applied to authenticity determination of an ID card printed using a reversible thermosensitive recording medium.

Description

Program, information processing method, and information processing system

[0001] The present disclosure relates to a program, an information processing method, and an information processing system, and in particular to a program, an information processing method, and an information processing system that enable appropriate authentication of printed materials printed using reversible thermosensitive recording media.

[0002] Technologies related to anti-counterfeiting measures for identification documents such as passports and ID (Identification) cards play an important role in preventing counterfeiting and imitation, as well as determining the authenticity of counterfeit products.

[0003] There are various anti-counterfeiting technologies available, including embedded IC chips, holograms, special inks that glow when exposed to ultraviolet light, and printing of tiny characters that are difficult to reproduce using a copy machine.

[0004] In addition to these anti-counterfeiting technologies, there is also authentication technology that uses images of the face of a bill taken with a camera on a smartphone or other device to determine its authenticity. This technology can be used not only for face-to-face identity verification but also for online identity verification conducted by financial institutions such as online banks.

[0005] As a technology for determining authenticity using an image of the face of an ID card, a technology has been proposed in which a concentric circular pattern is printed on the face of the ID card, and the authenticity of the ID card is determined based on whether or not the concentric circular pattern is visible in an image of the face of the ID card captured using a smartphone or other device (see Patent Document 1).

[0006] However, while the technology in Patent Document 1 can detect simple counterfeits, such as those made by attaching low-resolution printed materials to ID cards, it is possible to create ID cards with concentric circular patterns by printing an imitation pattern directly onto the card using a high-resolution printer, which may make it impossible to properly determine the authenticity of the card.

[0007] Therefore, as a countermeasure against counterfeits made using inkjet printers or sophisticated handmade counterfeits, a reversible thermosensitive recording medium has been developed in which information such as a person's name and photograph is printed on a plastic substrate using a laser (see Patent Document 2).

[0008] Since the text and photographs printed using reversible thermosensitive recording media cannot be easily altered, they are highly resistant to counterfeiting and are suitable for use in passports and ID cards.

[0009] JP 2023-130031 A International Publication No. 2023 / 176942

[0010] However, even if the characters and photographs printed using the reversible thermosensitive recording medium of Patent Document 2 are highly resistant to counterfeiting, if it is not possible to properly determine whether the printed characters and photographs are genuine works printed using a reversible thermosensitive recording medium, there is a risk that counterfeits printed by means other than printing using a reversible thermosensitive recording medium will be deemed genuine and used fraudulently.Similarly, there is a risk that genuine works printed using a reversible thermosensitive recording medium will be deemed genuine and their appropriate use will be hindered.

[0011] The present disclosure has been made in consideration of such circumstances, and in particular, is intended to enable the use of an image of the face of a ticket on which characters, photographs, etc. are printed to appropriately determine whether the printed characters or photographs are genuine works printed using a reversible thermosensitive recording medium, or counterfeit works printed using other methods.

[0012] A program and information processing system according to one aspect of the present disclosure are an image processing unit that extracts texture information from an image of the surface of a printed material captured at a predetermined resolution or higher, and a program that causes a computer to function as a judgment unit that judges whether the printed material has been printed using a predetermined printing method based on the texture information, and an information processing system that includes the image processing unit and the judgment unit.

[0013] An information processing method according to one aspect of the present disclosure is an information processing method that includes image processing that extracts texture information from an image of the surface of a printed matter captured at a predetermined resolution or higher, and a determination process that determines whether the printed matter has been printed using a predetermined printing method based on the texture information.

[0014] In one aspect of the present disclosure, texture information is extracted from an image of the surface of a printed material captured at a predetermined resolution or higher, and based on the texture information, it is determined whether the printed material has been printed using a predetermined printing method.

[0015] 10 is a diagram illustrating example images captured by a camera of a printed matter printed using a reversible thermosensitive recording medium and a printed matter printed using an inkjet printer. FIG. 11 is a diagram illustrating an example of a printed matter printed using a reversible thermosensitive recording medium. FIG. 12 is a diagram illustrating an example of an image captured of the printed matter of FIG. 2. FIG. 13 is a diagram illustrating an example of the spatial frequency spectrum of FIG. 3. FIG. 14 is a diagram illustrating an intensity ratio at a peak position in the image of FIG. 2. FIG. 15 is a diagram illustrating an example configuration of an information processing device of the present disclosure. FIG. 16 is a diagram illustrating a series of processing results by the information processing device of FIG. 6. FIG. 17 is a diagram illustrating a main angle. FIG. 18 is a diagram illustrating mask processing. FIG. 19 is a flowchart illustrating authenticity determination processing by the information processing device of FIG. 6. FIG. 19 is a flowchart illustrating preprocessing of FIG. 10. FIG. 19 is a flowchart illustrating frequency information extraction processing of FIG. 10. FIG. 20 is a flowchart illustrating score calculation processing of FIG. 21. FIG. 21 is a diagram illustrating a first application example of the present disclosure. FIG. 22 is a diagram illustrating a second application example of the present disclosure. FIG. 23 is a diagram illustrating an example configuration of a general-purpose computer.

[0016] Preferred embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant description will be omitted.

[0017] Hereinafter, embodiments of the present technology will be described in the following order: 1. Overview of the present disclosure 2. Preferred embodiment 3. First application example 4. Second application example 5. Example of execution by software

[0018] <<1. Overview of the Present Disclosure>> <Reversible Thermosensitive Recording Medium> The present disclosure particularly aims to enable, by using an image captured of a card face on which characters, photographs, etc. are printed, to appropriately determine whether the printed characters or photographs are genuine works printed using a reversible thermosensitive recording medium or counterfeit works printed using another method. Therefore, in describing the overview of the present disclosure, first, a description will be given of reversible thermosensitive recording media and printing using a reversible thermosensitive recording medium.

[0019] A reversible thermosensitive recording medium is, for example, a recording medium in which multiple reversible thermosensitive recording layers with different photothermal conversion wavelengths are laminated together via a heat-insulating layer. When pulsed with laser light of a predetermined wavelength, the reversible thermosensitive recording medium has the property that specific reversible thermosensitive recording layers selectively generate heat, and the generated heat can cause coloring or decoloring. By utilizing this property, information such as text and photographs can be printed (recorded) on the reversible thermosensitive recording medium, and the printed (recorded) information can be erased.

[0020] In printing technology using reversible thermosensitive recording media, objects such as letters and photographs are printed by irradiating the surface with pulses of laser light while scanning, and a unique texture with parallel lines spaced at equal intervals of approximately 50 to 150 μm appears on part or all of the printed surface. This feature can be captured by capturing an image of the surface of the printed surface with a camera with a certain magnification or higher.

[0021] Here, a "certain magnification or higher" is, for example, a resolution of 10 pixels / mm or higher, that is, a resolution at which a 10 mm square area is allocated to an image area of ​​100 pixels or more. For an image with 2000 square pixels, this is the magnification when an area of ​​about 20 cm square or less is captured.

[0022] Figure 1 compares images taken using the camera functions of two common smartphone models of a printed surface using a reversible thermosensitive recording medium and a printed surface using an inkjet printer. Both images capture a printed surface consisting of a photograph of the area around a person's eyes.

[0023] In Figure 1, the upper part of the figure is an image captured by camera A, the lower part is an image captured by camera B, the left side of the figure is an image captured of the printing surface using a reversible thermosensitive recording medium, and the right side of the figure is an image captured of the printing surface using an inkjet printer.

[0024] In both cameras A and B, a texture of linear lines arranged at equal intervals in the horizontal direction can be observed on the surface printed using the reversible thermosensitive recording medium, while a dot-like texture can be observed on the surface printed using the inkjet printer.

[0025] In the comparison in Figure 1, it was explained qualitatively that linear texture appears on the printed surface using a reversible thermosensitive recording medium. Below, we will explain quantitatively that linear texture appears on the printed surface using a reversible thermosensitive recording medium.

[0026] 2 shows an example of a printed matter (drawing) P1 formed by irradiating a reversible thermosensitive recording medium with a laser beam Lm. An image PA consisting of a person's face image formed by irradiating the laser beam Lm is visible on the surface of the printed matter P1. The image PA is formed by continuously irradiating multiple recording layers of the reversible thermosensitive recording medium with the laser beam Lm in the scanning direction, and is formed at a predetermined depth from the surface of the reversible thermosensitive recording medium.

[0027] 3 shows an example of an image P2 obtained by capturing a print (drawing) P1 (image PA). More specifically, image P2 in FIG. 3 is an enlarged image of the area within the rectangular dotted line frame (near the punctum) of image P1. The image data P2 includes a stripe pattern M of irregular width extending in the X direction, which is formed by irradiation with laser light Lm.

[0028] Therefore, the image PA of the printed matter (drawing) P1 includes a stripe pattern M of irregular widths extending in the X direction, which is formed by irradiation with the laser light Lm. In other words, the stripe pattern M of irregular widths extending in the X direction is drawn on the recording layer of the reversible thermosensitive recording medium as drawing marks caused by continuous irradiation of the laser light Lm on the surface of the reversible thermosensitive recording medium in the scanning direction. In this case, the stripe pattern M is formed by overlapping the colored portions of multiple recording layers.

[0029] If the image PA includes a person's face, the X direction (the direction in which the striped pattern M extends) may be parallel to the line segment connecting the eyes in the image. If the image PA includes a person's face, the Y direction (the direction perpendicular to the direction in which the striped pattern M extends) may be perpendicular to the line segment connecting the eyes in the image. In this case, the reversible thermosensitive recording medium may be rectangular in plan view, or may be a shape other than rectangular (e.g., polygonal, circular, elliptical, etc.) in plan view.

[0030] The image P2 is obtained by capturing an image using, for example, ring illumination. If the image is captured using general illumination (for example, simultaneous radiation), it becomes difficult to read the stripe pattern M included in the image PA from the image data due to light reflection on the surface of the reversible thermosensitive recording medium.

[0031] On the other hand, when ring illumination is used, the influence of light reflection on the surface of the reversible thermosensitive recording medium can be reduced, making it relatively easy to read the stripe pattern M included in the image PA from the image data.

[0032] Fig. 4 shows an example of the spatial frequency spectrum of image P2, in which the spatial frequency spectrum of image P2 in the X direction, represented by a solid waveform, and the spatial frequency spectrum of image P2 in the Y direction, represented by a dotted waveform, are superimposed on each other.

[0033] The spatial frequency spectrum of image P2 in the X direction contains one large peak Px, while the spatial frequency spectrum of image P2 in the Y direction contains periodic peaks Py that are not present in the spatial frequency spectrum of image P2 in the X direction.

[0034] Therefore, in the spatial frequency spectrum of the image data 21, a striped pattern M is drawn on the recording layer of the reversible thermosensitive recording medium in such a manner that the profile in the Y direction, which is perpendicular to the X direction, has a periodic peak Py that is not present in the profile in the X direction.

[0035] The left part of Fig. 5 shows an example of measurement areas specified in image P2, including the vicinity of the lacrimal punctum (measurement area a), the vicinity of the cheek boundary (measurement area b), a background without a person (measurement area c), the vicinity of the corner of the eye (measurement area d), and the vicinity of the lip edge (measurement area e).

[0036] The right part of Fig. 5 shows an example of the peak ratio in each measurement region. The peak ratio means the intensity ratio at the position of peak Py when the spatial frequency spectrum in the X direction and the spatial frequency spectrum in the Y direction of image data P2 are superimposed on each other.

[0037] From the right side of FIG. 5, it can be said that a striped pattern M is drawn on the multiple recording layers of the reversible thermosensitive recording medium in such a manner that the intensity ratio satisfies the following formula:

[0038] S2 / S1≧1.2 S1: intensity at the position of peak Py of the profile in the X direction S2: intensity at the position of peak Py of the profile in the Y direction S2 / S1: the intensity ratio

[0039] It can be said that an image PA including a striped pattern M having the quantitative characteristics described above is printed as a drawing P1 on the reversible thermosensitive recording medium.

[0040] Therefore, in the present disclosure, an image of the face of a card on which characters, photographs, etc. are printed is captured, and it is determined whether or not this stripe pattern M can be observed in the captured image.

[0041] As a result, if the striped pattern M can be observed, it can be determined that the printed characters and photographs are genuine and recorded on a reversible thermosensitive recording medium, and conversely, if the striped pattern M cannot be observed, it can be determined that the printed work is a counterfeit printed by another method.

[0042] As a result, it is possible to properly perform authenticity determination by using images of characters and photographs printed on the face of a ticket to determine whether the printed characters and photographs are genuine works recorded on a reversible thermosensitive recording medium.

[0043] <<2. Preferred embodiment>> Next, with reference to the block diagram in FIG. 6 , an example configuration of an information processing device that can appropriately perform authenticity determination, that is, determine whether a printed matter is an authentic work recorded on a reversible thermosensitive recording medium, from an image captured of the printed matter to which the technology of the present disclosure is applied.

[0044] The information processing device 31 in FIG. 6 is configured to be a smartphone or a tablet terminal, and is assumed to be equipped with at least a camera capable of capturing images.

[0045] The information processing device 31 is composed of a control unit 41, an input unit 42, an output unit 43, a memory unit 44, a communication unit 45, a drive 46, a removable storage medium 47, and a camera 50, which are connected to each other via a bus 48 and can send and receive data and programs.

[0046] The control unit 41 is composed of a processor and a memory, and controls the overall operation of the information processing device 31. The control unit 41 also includes an image extraction unit 51, a preprocessing unit 52, a frequency information extraction unit 53, a score calculation unit 54, and an authenticity determination unit 55.

[0047] The image extraction unit 51 extracts an image of a predetermined region to be determined from within the image to be determined for authenticity, and outputs the extracted image to the preprocessing unit 52. For example, if the image to be determined for authenticity is a photograph of a person's face, the predetermined region to be extracted is a specific region such as the area around the eyes, nose, and lips. Furthermore, the image to be determined for authenticity is not limited to a photograph of a person's face, but may also be an area on which other text or a photograph is printed.

[0048] In order to adaptively adjust the sharpness of the original image supplied from the image extraction unit 51, the pre-processing unit 52 performs filtering using, for example, an LPF (Low Pass Filter) consisting of a Gaussian filter, obtains a difference image from the original image, and further performs normalization based on variation so that the texture corresponding to the above-mentioned stripe pattern M is emphasized, and outputs the result to the frequency information extraction unit 53.

[0049] The frequency information extraction unit 53 applies a two-dimensional FFT (Fast Fourier Transformation) to the image with texture emphasis that is the preprocessing result of the preprocessing unit 52, performs frequency analysis of the texture, and outputs the result of the frequency analysis to the score calculation unit 54.

[0050] Based on the results of the frequency analysis, the score calculation unit 54 calculates a score that becomes large when a texture corresponding to the striped pattern M described above is observed, and outputs the score to the authenticity determination unit 55 .

[0051] The authenticity determination unit 55 determines whether the image to be determined for authenticity is a drawing printed using a reversible thermosensitive recording medium based on whether the score calculated by the score calculation unit 54 is greater than a threshold value.

[0052] The operations of the image extraction unit 51, preprocessing unit 52, frequency information extraction unit 53, score calculation unit 54, and authenticity determination unit 55 will be described in detail below with reference to FIG. 7 along with an explanation of the processing.

[0053] The input unit 42 is composed of input devices such as a keyboard, a mouse, and a touch panel through which the user inputs operation commands, and supplies various input signals to the control unit 41 .

[0054] The output unit 43 is controlled by the control unit 41 and includes a display unit and an audio output unit. The output unit 43 outputs and displays images of operation screens and processing results on the display unit, which is made up of a display device such as an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) display. The output unit 43 also controls the audio output unit, which is made up of an audio output device, to play various sounds, music, sound effects, and the like.

[0055] The storage unit 44 is composed of a hard disk drive (HDD), a solid state drive (SSD), or a semiconductor memory, and is controlled by the control unit 41 to write or read various data and programs.

[0056] The communication unit 45 is controlled by the control unit 41 and realizes communication via wired or wireless means, such as LAN (Local Area Network) or Bluetooth (registered trademark), and transmits and receives various data and programs to and from various devices via the network as necessary.

[0057] The drive 46 reads and writes data from and to a removable storage medium 47 such as a magnetic disk (including a flexible disk), an optical disk (including a CD-ROM (Compact Disc-Read Only Memory) and a DVD (Digital Versatile Disc)), a magneto-optical disk (including an MD (Mini Disc)), or a semiconductor memory.

[0058] The camera 50 has a resolution of at least 10 pixels / mm, i.e., a certain magnification such that a 10 mm square area is allocated to an image area of ​​at least 100 pixels square, and supplies the captured image to the control unit 41.

[0059] <Flow of each component realized by the information processing device and authenticity determination processing> Next, referring to Figure 7, when using an image P11 captured of a printed matter (drawing) printed using a reversible thermosensitive recording medium and an image P21 captured of a printed matter (drawing) printed using an inkjet printer, the functions realized by each component of the information processing device 31 in Figure 6 will be explained together with the processing results realized by each function.

[0060] (Preprocessing) In order to determine whether a printed matter was printed using a reversible thermosensitive recording medium based on whether the printed matter has a unique texture, based on an image of the surface of the printed matter, it is necessary to reduce the influence of the printed content (printed matter) and acquire texture information with emphasis. For this reason, preprocessing is performed to emphasize the texture using the following procedure.

[0061] First, the image extraction unit 51 extracts, as an image X, a recognition target area to be applied to authenticity determination from the recognition target image Pin, for example.

[0062] More specifically, when the recognition target image Pin is, for example, the images P11 and P21 in Fig. 7, the image extraction unit 51 extracts, for example, images P12 and P22 from the images P11 and P21 in Fig. 7 as the image X consisting of the recognition target region to be applied to the authenticity determination. Note that Fig. 7 shows an example in which the vicinity of the eyes of each of the images P12 and P22 is extracted as the recognition target region, but this is not limiting, and other regions may be set as the recognition target region.

[0063] Next, a filter process is required to adaptively adjust the sharpness of the image for each region. Therefore, the pre-processing unit 52 applies a low pass filter (LPF) to the image X using the following equation (1) to obtain a smoothed image A, and generates an image X' that is the difference between the image X and the smoothed image A.

[0064] X'=X-A A=LPF(X)...(1)

[0065] Here, LPF(U) is a function that applies a smoothing filter such as a Gaussian filter to each pixel of image U. Note that the smoothing filter may be other than an LPF, and may be, for example, a contrast limited adaptive histogram equalization (CLAHE).

[0066] Next, the pre-processing unit 52 calculates the following equation (2) to generate an image X″ for each region of the image X′ of the recognition target region by normalizing each pixel of the image X′ according to the variation of the surrounding pixels.

[0067] X''=ADD(MUL(DIV(X',s),C),D)...(2)

[0068] In the above equation, DIV(L, M) is a function that divides image L by image M for each pixel, MUL(N, O) is a function that multiplies each pixel of image N by a constant O, and ADD(P, Q) is a function that adds a constant Q to each pixel of image P. Furthermore, S is the variation in pixel values ​​of neighboring pixels for each pixel of image X'. Here, the variation S is calculated, for example, as in the following equation (3): S=SQRT(LPF(SQUARE(X'))) (3)

[0069] Here, SQRT(J) is a function that calculates and outputs the square root of each pixel of image J, LPF(K) is a function that applies a smoothing filter such as a Gaussian filter to each pixel of image K, and SQUARE(L) is a function that calculates and outputs the square of each pixel of image L.

[0070] That is, equation (2) expresses the normalization of the pixel value of each pixel of image X' using the variation S (standard deviation) for each pixel of image X' calculated by equation (3). As a result, each pixel of image X'' has a relatively small pixel value if the variation S of its surrounding pixels is large, and a relatively large pixel value if the variation S of its surrounding pixels is small.

[0071] As a result, the global information of the captured image X' is reduced, while an image X'' is generated in which the local texture is emphasized. In other words, by performing the calculation using equation (2), the distribution of the surrounding pixel values ​​for each pixel of image X' is made uniform, so that the sharpness of image X'' for each region becomes consistent regardless of the content.

[0072] More specifically, the pre-processing unit 52 generates images P13 and P23 of the eyes in Figure 7 corresponding to image X'' by, for example, applying pre-processing using the calculations described with reference to equations (1) to (3) above to images P12 and P22 in Figure 7 corresponding to image X'.

[0073] In images P13 and P23, it can be seen that the information derived from the content is weakened as a result of the local uniformity of the surrounding areas, both in areas with strong contrast (e.g., around the eyelashes) and areas with weak contrast (e.g., the lower eyelid).On the other hand, in images P13 and P23, it can be seen that the original texture is emphasized in areas where texture is present.

[0074] That is, in image P13, the striped texture that is unique to prints made using reversible thermosensitive recording media is emphasized, while in image P23, the dot-like texture that is unique to prints made using an inkjet printer is emphasized.

[0075] (Frequency information extraction) In the present disclosure, when determining whether a printed matter is made using a reversible thermosensitive recording medium, it is determined whether the printed matter has a texture unique to printed matter made using a reversible thermosensitive recording medium, in which "the printed matter has parallel lines arranged at equal intervals," based on an image captured of the surface of the printed matter.

[0076] To make this determination, a method is required to quantify the likelihood of the appearance of these "equally spaced parallel lines." More specifically, it is necessary to distinguish between "equally spaced parallel lines," a texture unique to prints made using reversible thermosensitive recording media, and other textures. Examples of other textures include "equally spaced dots" that are observed in prints made using an inkjet printer.

[0077] However, simply detecting peaks in the frequency spectrum of an FFT (Fast Fourier Transform) as in Patent Document 1 makes it difficult to distinguish between parallel lines and dots because the peaks are similar.

[0078] Therefore, in this disclosure, a distinction is made between a texture consisting of "equally spaced parallel lines" and a texture consisting of "equally spaced dots" based on whether the angle at which the peak frequency is observed is in one direction or multiple directions.

[0079] More specifically, the frequency information extraction unit 53 applies a two-dimensional FFT to the image X″ that has been subjected to the above-described preprocessing, generates a power spectrum image B from the result of the FFT, and then performs polar coordinate transformation to output a frequency analysis image Y.

[0080] More specifically, the frequency information extraction unit 53 performs a two-dimensional FFT on images P13 and P23 corresponding to the preprocessed image X'', and generates images P14 and P15 corresponding to the power spectrum image B from the results of the two-dimensional FFT. Furthermore, the frequency information extraction unit 53 performs polar coordinate transformation on the images P14 and P24 corresponding to the power spectrum image B, thereby generating images P15 and P25 corresponding to the frequency analysis image Y.

[0081] The images P14 and P24, which are power spectrum image B, each show a spectral distribution in which the minimum frequency is at the center, and the frequency increases toward the periphery, and the greater the power, the brighter the spectrum becomes. The concentric circles shown by dotted lines in images P14 and P24 in Figure 7 represent the distribution of positions with the same frequency f. In addition, in power spectrum image B, the angle θ at which the frequency f is detected relative to the center position is expressed as the angle between the center position and a horizontal line in the figure.

[0082] The angle θ at which frequency f is detected is as follows: For example, as shown in image P31 in Fig. 8 , when "equally spaced parallel lines" are represented by parallel lines sloping upward to the right as indicated by the dotted lines, and the spacing between the "equally spaced parallel lines" is detected as frequency f, frequency f is detected in the direction of the dashed-dotted arrow. Therefore, in power spectrum image B in Fig. 8 , the angle θ at which frequency f is detected is expressed as angle θf.

[0083] 8, when "equally spaced parallel lines" are obtained as a texture, only one angle is detected as the predetermined frequency f, but in the case of a dot-shaped texture, various frequencies f are detected at multiple angles. In this disclosure, attention is paid to this point to determine whether printing is performed using a reversible thermosensitive recording medium or an inkjet printer.

[0084] Furthermore, images P15 and P25 are images obtained by polar coordinate transformation of images P14 and P24, respectively.

[0085] The reason for introducing polar coordinate transformation is to simplify the calculation in the subsequent score calculation process by representing the axis representing frequency (or wavelength) and the axis representing angle perpendicularly in the vertical and horizontal directions, as shown in images P15 and P25 in Figure 7. However, since the purpose is to simplify the calculation, it is not necessarily necessary to introduce it. The score calculation process will be described in detail later.

[0086] (Score calculation) The score calculation unit 54 applies a masking process to the frequency analysis image Y to generate a frequency analysis image Y' in order to limit the area to only those areas where specific frequencies are likely to be observed when printing using a reversible thermosensitive recording medium, and calculates a percentage score based on the frequency analysis image Y'.

[0087] More specifically, the score calculation unit 54 masks at least one of the low-frequency band below a predetermined value and the high-frequency band above a predetermined value in the frequency analysis image Y. Furthermore, for the angle range of 0° to 360°, there is symmetry between the 0° to 179° range and the 180° to 360° range, resulting in redundant information. Therefore, the score calculation unit 54 masks either the 0° to 179° range or the 180° to 360° range and uses the other range for calculation. Therefore, for example, the score calculation unit 54 uses only the 0° to 179° range for calculation and masks the 180° to 360° range. That is, the score calculation unit 54 generates the frequency analysis image Y' by masking the frequency analysis image Y. For example, in the case of images P15 and P25 in FIG. 7 , the score calculation unit 54 performs masking processing to generate images P16 and P26.

[0088] That is, for example, when image P51 in FIG. 9 is supplied as the frequency analysis image Y, the score calculation unit 54 generates image P52 by masking the low-frequency band and the high-frequency band of image P51. Furthermore, the score calculation unit 54 generates image P53 by masking the range of 180° to 360° of image P52. Note that, since the masking process only requires masking either the range of 0° to 179° or the range of 180° to 360°, it is also possible to mask the range of 0° to 179° and use the range of 180° to 360° for calculation. Furthermore, the masking process is a process for simplifying calculation and is not essential.

[0089] Next, the score calculation unit 54 uses the frequency analysis image Y' to calculate the main angle θ at which the frequency f is detected by calculating the following equation (4).

[0090] θ=ARGMAX(SUMCOL(Y'))...(4)

[0091] Here, SUMCOL(G) is a function that adds values ​​in the column direction (frequency direction) of image G for each row of image G (for each angle θ), and ARGMAX(H) is a function that calculates the maximum value of H, that is, in this case, the angle θ that is the maximum value among the results of adding values ​​in the column direction (frequency direction), as the main angle θ.

[0092] The score calculation unit 54 also calculates the principal frequency f at the principal angle θ. The method of calculating the principal frequency f at the principal angle θ is, for example, to adopt the frequency f at which the power is maximum among the principal angles θ of the frequency analysis image Y′ as the principal frequency f at the principal angle θ.

[0093] Furthermore, the score calculation unit 54 calculates the proportion of the power of the main angle θ at the main frequency f to the sum of the power of all angles 0 to 179°, that is, the contribution rate of the main angle θ, as a proportion score.

[0094] The method for calculating the proportion score of the principal angle θ at the principal frequency f is, for example, to use the proportion of the power that the principal angle θ accounts for to the total power of 180 elements from 0 to 179° of the principal frequency f of image Y', i.e., the contribution rate of the principal angle θ.

[0095] In the case of dot patterns observed in inkjet printing, there are multiple angles at which peaks appear in the power spectrum image, so the percentage score calculated by the above process is low because the contribution rate of the main angle is dispersed across multiple angles.

[0096] On the other hand, in the case of a pattern of evenly spaced parallel lines observed in printing using a reversible thermosensitive recording medium, there is generally only one angle at which a peak occurs in the power spectrum image, and therefore the contribution rate of the main angle is concentrated at that one angle, resulting in a high percentage score.

[0097] As a result, the ratio score functions as a score that separates textures that have peaks at multiple angles, such as inkjet prints, from textures that have only one peak, such as prints made using reversible thermosensitive recording media.

[0098] Furthermore, by adopting the "contribution rate of the main angle θ at the main frequency f," it becomes possible to flexibly respond to changes in the main frequency f due to changes in the imaging distance or magnification, and changes in the main angle θ due to changes in the imaging direction or angle.

[0099] As a result, it is possible to achieve appropriate authentication even if the imaging angle or the distance to the subject changes.

[0100] To summarize the above processing, according to the present disclosure, "detailed texture is captured at a certain magnification or higher," "the influence of the print content (low frequency information) is reduced by adaptive filtering to extract texture information (high frequency information)," and then "whether or not the print has characteristics specific to prints made using reversible thermosensitive recording media (the angle at which the main frequency appears is in a single direction)" is determined to determine whether or not the print is made using reversible thermosensitive recording media.

[0101] In other words, the above method makes it possible to appropriately determine whether a printed matter is a printed matter using a reversible thermosensitive recording medium, based only on an image of the face of the card.

[0102] In addition, adaptive filtering reduces the influence of printed content (low-frequency information) and emphasizes texture information (high-frequency information), making it possible to achieve highly accurate and stable authenticity determination even when any content is printed, without limiting the printed content to concentric circular patterns, etc.

[0103] Furthermore, the percentage score, which is the "contribution rate of the main angle θ at the main frequency f" related to the authenticity determination, is less dependent on the angle at which the parallel lines appear or the scale, making it possible to achieve stable authenticity determination without being affected by changes in the imaging angle or the distance to the subject.

[0104] <Authenticity Determination Process> Next, the authenticity determination process of the printed matter on the face of the card by the information processing device 31 in FIG. 6 will be described with reference to the flowchart in FIG.

[0105] In step S31 , the camera 50 captures an image of a printed matter such as a banknote face that is the target of authentication, and supplies the captured image Pin to the control unit 41 .

[0106] In step S32, the image extraction unit 51 extracts and cuts out, as image X, a recognition target area to be applied to authenticity determination from the recognition target image Pin.

[0107] In step S33, the preprocessing unit 52 performs preprocessing on the image X to generate an image X'' in which the texture is enhanced, and supplies the image X'' to the frequency information extraction unit 53. Details of the preprocessing will be described later with reference to the flowchart in FIG. 11.

[0108] In step S34, the frequency information extraction unit 53 performs frequency information extraction processing on the image X'' to generate a frequency analysis image Y, and supplies the frequency analysis image Y to the score calculation unit 54. The frequency information analysis processing will be described in detail later with reference to the flowchart of FIG.

[0109] In step S35, the score calculation unit 54 executes a score calculation process based on the frequency analysis image Y to calculate the above-mentioned proportion score and output it to the authenticity determination unit 55. Details of the score calculation process will be described later with reference to the flowchart in FIG.

[0110] In step S36, the authenticity determination unit 55 determines whether the ratio score is higher than a predetermined threshold value.

[0111] If it is determined in step S36 that the percentage score is higher than the predetermined threshold, the process proceeds to step S37.

[0112] In step S37, the authenticity determination unit 55 determines that the print imaged by the camera 50 is an authentic work printed using a reversible thermosensitive recording medium.

[0113] Also, if it is determined in step S36 that the percentage score is not higher than the predetermined threshold, the process proceeds to step S38.

[0114] In step S38, the authenticity determination unit 55 determines that the print imaged by the camera 50 is not printed using a reversible thermosensitive recording medium and is a counterfeit.

[0115] By performing the above process, the printed matter to be determined for authenticity is photographed with the camera 50, and by simply using the photographed image, it is possible to properly determine whether the printed matter is an authentic work printed using a reversible thermosensitive recording medium.

[0116] <Preprocessing> Next, the preprocessing by the preprocessing unit 52 will be described with reference to the flowchart of FIG.

[0117] In step S51, the preprocessing unit 52 acquires an image X.

[0118] In step S52, the pre-processing unit 52 generates a smoothed image A by applying an LPF to each pixel of the image X as adaptive filtering.

[0119] In step S53, the pre-processing unit 52 generates an image X' (=X-A) ​​by subtracting the image A from the image X using the above-mentioned equation (1).

[0120] In step S54, the pre-processing unit 52 calculates the variation S, which is the standard deviation of the pixel values ​​of the surrounding pixels for each pixel of the image X', by using the above-mentioned equation (3).

[0121] In step S55, the pre-processing unit 52 normalizes the image X' with the variation S by calculation using the above-mentioned equation (2), and calculates the normalized result as an image X'' with enhanced texture.

[0122] In step S56, the preprocessing unit 52 outputs the generated image X'' with the texture enhanced to the frequency information extraction unit 53.

[0123] Through the above processing, image X, which is the subject of authenticity determination, is smoothed to generate image A, image A is subtracted from image X, the original image, to generate image X', and image X' is normalized by pixel-by-pixel variation S to generate image X'' with enhanced texture.

[0124] More specifically, by the above processing, for example, as described with reference to FIG. 7, preprocessing is performed on images P12 and P22 corresponding to image X, the authenticity of which is to be determined, to generate images P13 and P23 corresponding to image X". This makes it possible to reduce the influence of the content to be captured and to apply preprocessing so as to create images with enhanced texture.

[0125] <Frequency Analysis Processing> Next, the frequency analysis processing will be described with reference to the flowchart of FIG.

[0126] In step S71, the frequency information extraction unit 53 performs a two-dimensional FFT on the image X''.

[0127] In step S72, the frequency information extraction unit 53 converts the result of the two-dimensional FFT on the image X'' into a power spectrum image B.

[0128] In step S73, the frequency information extraction unit 53 performs polar coordinate transformation on the power spectrum image B to generate a frequency analysis image Y, and outputs it to the score calculation unit .

[0129] By the above processing, for example, as described with reference to Fig. 7, images P13 and P23 corresponding to image X'' are subjected to a two-dimensional FFT, and the results of the two-dimensional FFT are converted into power spectrum images B such as images P14 and P24 in Fig. 7. Then, power spectrum image B is subjected to polar coordinate transformation to generate frequency analysis images Y such as images P15 and P25.

[0130] <Score Calculation Process> Next, the score calculation process will be described with reference to the flowchart of FIG.

[0131] In step S91, the score calculation unit 54 acquires a frequency analysis image Y.

[0132] In step S92, the score calculation unit 54 performs mask processing on the frequency analysis image Y as described with reference to FIG. 9 to generate a frequency analysis image Y'.

[0133] In step S93, the score calculation unit 54 calculates the main angle θ based on the frequency analysis image Y′ by performing the calculation shown in equation (4) above.

[0134] In step S94, the score calculation unit 54 calculates the frequency f that is the maximum value among the principal angles θ of the frequency analysis image Y′ as the principal frequency f of the principal angle θ.

[0135] In step S95, the score calculation unit 54 calculates the ratio of the power of the main angle θ to the total power of angles 0 to 179° for the main frequency f of the main angle θ as a ratio score.

[0136] Through the above processing, based on the frequency analysis image Y', only the direction in which the "equally spaced parallel lines" that are a texture unique to printed materials using reversible thermosensitive recording media are arranged in the power spectrum image, i.e., only one main angle θ, is detected, so the percentage score is not dispersed by other angles θ.

[0137] On the other hand, in the case of a printed material other than a printed material using a reversible thermosensitive recording medium, for example, when there are multiple angles similar to the main angle θ, such as a dot-like texture, the percentage score will be smaller as it will be dispersed by the other angles.

[0138] As a result, when the percentage score is higher than a predetermined value, it can be considered that "equally spaced parallel lines," a texture unique to printed matter using reversible thermosensitive recording media, have been detected, making it possible to determine that the printed matter is made using reversible thermosensitive recording media.

[0139] <<3. First Application Example>> The above has described an example in which the information processing device 31 is a smartphone, a tablet terminal, or the like, and all processes from capturing an image of a printed material to determining the authenticity of the captured printed material are performed in the information processing device 31. In other words, the above has described an example in which a user who owns a smartphone or a tablet terminal determines the authenticity of an ID card or the like and verifies the identity of the person who owns the ID card.

[0140] However, the present invention may also be applied to face-to-face identity verification using information such as characters and photographs printed on a reversible thermosensitive recording medium. That is, when an inspector performs face-to-face identity verification by visually checking information such as characters and photographs printed on an ID card or the like made of a reversible thermosensitive recording medium, a camera may be installed at the inspection site to capture an image of the ID card face, and the above-described technology of the present disclosure may be applied to determine the authenticity of the ID card and realize a process to assist the inspector in identity verification.

[0141] Figure 14 shows an example configuration of an information processing system that applies the technology disclosed herein to assist inspectors in cases where they conduct face-to-face identity verification based on information such as text and photographs printed on presented ID cards, etc., by determining the authenticity of the ID card and presenting the determination results.

[0142] The information processing system 101 in FIG. 14 is composed of an information processing device 31' and a camera 50'.

[0143] The information processing device 31' has the same basic functions as the information processing device 31, but further has a camera 50' connected to it as a separate device. Since the camera 50' is connected to it as a separate device, the camera 50 is no longer an essential component of the information processing device 31'.

[0144] The camera 50' has the same basic functions as the camera 50, but is provided separately from the information processing device 31' and is used by an inspector when verifying the identity of a person in person. That is, the inspector, for example, holds an ID card presented by a person to be verified in person over the camera 50' to capture an image, and then supplies the captured image to the information processing device 31'. Note that while Figure 14 shows examples of driver's licenses 111-1 and passports 111-2 as examples of ID cards, other ID cards may be used as long as the authentic ID card is a printed matter using a reversible thermosensitive recording medium.

[0145] With this configuration, the information processing device 31' can assist the inspector by performing the above-mentioned authenticity determination process based on an image of the face of an ID card or the like captured by the camera 50' and presenting the determination result to the inspector.

[0146] The information presented to the inspector may be only the result of the authenticity determination, or, for example, an image captured by camera 50' may also be presented, or other intermediate images as described with reference to FIG. 7 may be presented.

[0147] In addition, instead of the camera 50', a general smartphone or tablet terminal may be used, and by using only the camera function of each, the face of the ID card may be captured and the captured image may be supplied to the information processing device 31'.

[0148] <<4. Second Application Example>> In the above, an example has been described in which information such as characters and photographs printed on a reversible thermosensitive recording medium is applied to face-to-face identity verification work.

[0149] However, it may also be applied to identity verification via a network. That is, when an image of the face of an ID card or the like made of a reversible thermosensitive recording medium or the like is submitted via a network, the authenticity of the ID card may be determined based on the submitted image by applying the technology disclosed herein, thereby realizing online identity verification.

[0150] Figure 15 shows an example configuration of an information processing system that, when information is submitted via a network, capturing an image of the face of an ID card, such as text or a photograph printed on the card, made of a reversible thermosensitive recording medium, etc., the authenticity of the ID card is determined based on the submitted image by applying the technology disclosed above.

[0151] The information processing system 101 ″ in FIG. 15 is composed of an information processing device 31 ″, a camera 50 ″, and the information processing device 31 .

[0152] Camera 50'' basically has the same functions as cameras 50 and 50', and has the same function of capturing images of the face of an ID card, etc., but also transmits the captured image of the face of the card to information processing device 31'' via network 121.

[0153] Furthermore, the information processing device 31 here does not perform authentication, but captures an image of the face of the ID card or the like, and transmits the captured image to the information processing device 31 ″ via the network 121 .

[0154] The information processing device 31'' is composed of a cloud server, a network server, etc., and has the same basic functions as the information processing device 31, but it accepts an image of the face of the ID card or other card to be judged from the camera 50'' or the information processing device 31 via the network 121, performs the above-mentioned authenticity judgment process, and then performs identity verification.

[0155] In the information processing system 101'' of FIG. 15, the information processing device 31 can capture an image of the face of the ID card or the like to be judged and perform the authenticity determination process based on the captured image, so it may be configured to transmit only its own authenticity determination result to the information processing device 31'' via the network 121. In this case, the information processing device 31'' may obtain only the authenticity determination result of the information processing device 31 to realize identity verification.

[0156] In addition, instead of transmitting the authenticity determination result to the information processing device 31'', the information processing device 31 may, for example, transmit the intermediate image described with reference to Figure 7 together with the captured image, and the information processing device 31'' may perform the authenticity determination process using the intermediate image.

[0157] (First Modification (Use of Camera Information)) Authenticity determination processing may be performed by combining the captured image with the angle of view and focus information of the camera used for capturing the image. For example, the imaging range may be calculated from the angle of view and focus information of the camera used for capturing the image, and if the difference from the previously estimated imaging range is greater than a predetermined value, the percentage score determined from the image may be adjusted lower.

[0158] For example, if an image of 20 to 100 mm square is assumed to be taken, but an image of 500 m square is actually taken, there is a possibility that a large-scale print with a striped pattern similar to that printed using a reversible thermosensitive recording medium could be counterfeited, so the percentage score is adjusted lower to take into account the possibility of counterfeiting.

[0159] By introducing this process, it becomes possible to prevent fraud using large-scale prints that have striped patterns similar to those printed using reversible thermosensitive recording media.

[0160] (Second Modification (Regarding Recognition Target Area)) In this specification, the recognition target area has been described using the area around the eyes as an example, but the recognition target area is not limited to the eyes. For example, it may be other facial features such as the nose or mouth, or even if it is an eye, it may be further limited to using only the iris area without using the area around the eye. Furthermore, instead of a face, characters, logos, QR codes (registered trademark), etc. may be detected and used to determine authenticity.

[0161] By setting various recognition target areas in this way, when recognizing a face photo on an ID card, if visual authentication fails, it is possible to re-evaluate using a different part, or to improve the accuracy of the judgment by judging using multiple parts.

[0162] (Third variant (regarding the use of multiple types of adaptive filter processing)) The processing performed by the pre-processing unit 52 in Figure 2 may use multiple types of adaptive filter processing, and the final percentage score may be calculated by combining (averaging, maximizing, etc.) the multiple percentage scores obtained after performing subsequent processing on each output.

[0163] The multiple types of adaptive filtering include, for example, using LPFs with different kernel sizes, and using adaptive contrast adjustment processing such as CLAHE in combination with the processing of the pre-processing unit 52 .

[0164] When N types of adaptive filter processing are used, N percentage scores are calculated, and the final percentage score is calculated by using the maximum, average, or minimum value of the N scores and aggregating them into a single percentage score.

[0165] In addition, instead of consolidating the processes for calculating the ratio score into one, after a judgment process based on comparison with a predetermined threshold, N authenticity judgment results may be obtained, and then the N results may be consolidated into one authenticity judgment result by majority vote.

[0166] By using multiple types of adaptive filtering in this way, it is possible to improve robustness against changes in the content and scale of the image.

[0167] (Fourth Variant Example (Regarding Use of Multiple Image Channels)) The processing performed by the pre-processing unit 52 may involve applying adaptive filter processing to each of multiple image channels, performing subsequent processing on each output, and then combining (averaging, maximizing, etc.) the multiple judgment scores obtained to calculate a final score.

[0168] In the case of an RGB color image, the multiple image channels can be three types: R channel, G channel, and B channel. When M types of adaptive filter processing are used, M percentage scores are calculated, and the final percentage score is calculated by combining the maximum, average, minimum, etc. of the M percentage scores into a single percentage score.

[0169] In addition, instead of consolidating the scores into one during the score calculation process, M authenticity determination results may be calculated after a predetermined judgment based on a comparison between the percentage score and a predetermined threshold, and then the M authenticity determination results may be consolidated into one authenticity determination result by majority vote.

[0170] By combining the determination results from multiple channels in this way, it becomes possible to obtain more stable determination results.

[0171] However, for an input RGB image, instead of processing the R, G, and B channels independently, the RGB may be converted to grayscale in advance and aggregated into a single channel before being applied to subsequent processing.

[0172] (Fifth Variant Example (Regarding Use of Images Acquired at Multiple Times)) The processing performed by the pre-processing unit 52 may involve applying adaptive filter processing to each of multiple images acquired at multiple times, and then performing subsequent processing based on the outputs of each to combine (average, maximum, etc.) the multiple percentage scores obtained after the processing to calculate a final percentage score. Examples of images acquired at multiple times include videos in which images are captured in chronological order. By combining authenticity determination results based on imaging results at multiple times in this way, it is possible to obtain more stable authenticity determination results.

[0173] (Sixth Modification (Use of DNN)) A deep neural network (DNN) may be used to perform some or all of the processing of the preprocessing unit 52, the frequency information extraction unit 53, and the score calculation unit 54. For example, a machine-learned DNN may be used to determine whether an input image is an image of a printed matter using a reversible thermosensitive recording medium, and the authenticity determination process may be performed by comparing the output score of the DNN with a predetermined threshold value instead of using a percentage score.

[0174] (Seventh Modification (Regarding Result Feedback)) The result output may be fed back not only as a percentage score or an authenticity determination result, but also as error information obtained in the process.

[0175] For example, in the process of extracting and cutting out the recognition target area, if the eye area is the target and the eye area is not detected, the subsequent processing may not be performed and an error message stating that "the recognition target area (eye area) was not detected" may be displayed.

[0176] Also, for example, if the result of frequency information extraction is that the value of the power spectrum of the frequency information is lower than a predetermined value, an error message such as "The image may be out of focus" may be displayed.

[0177] This feedback reduces the number of cases where authenticity determination is not performed correctly due to user operation errors, and shortens the time it takes for the user to perform a correct authenticity determination.

[0178] (Eighth Variant Example (Regarding Applicable Objects)) The present disclosure is a technology for determining whether the texture of a target object is a predefined printed matter, and therefore, in addition to determining the authenticity of ID cards and passports as described above, it can also be applied to determining the authenticity of trading cards and medicine packages printed with printed matter using reversible thermosensitive recording media.

[0179] This makes it possible to confirm that trading cards are not illegally counterfeit and to identify illegally produced medicines by verifying that the packages are genuine, thereby preventing the distribution of illegal trading cards and illegal medicines.

[0180] <<5. Example of Execution by Software>> The above-described series of processes can be executed by hardware, but can also be executed by software. When the series of processes is executed by software, the program constituting the software is installed from a recording medium into a computer incorporated in dedicated hardware, or into, for example, a general-purpose computer that can execute various functions by installing various programs.

[0181] 16 shows an example of the configuration of a general-purpose computer. This computer has a built-in CPU (Central Processing Unit) 1001. An input / output interface 1005 is connected to the CPU 1001 via a bus 1004. A ROM (Read Only Memory) 1002 and a RAM (Random Access Memory) 1003 are connected to the bus 1004.

[0182] The input / output interface 1005 is connected to an input unit 1006 including input devices such as a keyboard and a mouse through which a user inputs operation commands, an output unit 1007 that outputs a processing operation screen and images of processing results to a display device, a storage unit 1008 including a hard disk drive or the like that stores programs and various data, and a communication unit 1009 including a LAN (Local Area Network) adapter or the like that executes communication processing via a network typified by the Internet. Also connected is a drive 1010 that reads and writes data from / to a removable storage medium 1011 such as a magnetic disk (including a flexible disk), an optical disk (including a CD-ROM (Compact Disc-Read Only Memory) and a DVD (Digital Versatile Disc)), a magneto-optical disk (including an MD (Mini Disc)), or a semiconductor memory.

[0183] The CPU 1001 executes various processes in accordance with a program stored in a ROM 1002 or a program read from a removable storage medium 1011 such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, installed in a storage unit 1008, and loaded from the storage unit 1008 into a RAM 1003. The RAM 1003 also stores data necessary for the CPU 1001 to execute various processes as appropriate.

[0184] In a computer configured as described above, the CPU 1001 performs the above-described series of processes by, for example, loading a program stored in the memory unit 1008 into the RAM 1003 via the input / output interface 1005 and the bus 1004 and executing it.

[0185] The program executed by the computer (CPU 1001) can be provided by being recorded on a removable storage medium 1011 such as a package medium, for example. The program can also be provided via a wired or wireless transmission medium such as a local area network, the Internet, or digital satellite broadcasting.

[0186] In a computer, a program can be installed in the storage unit 1008 via the input / output interface 1005 by inserting a removable storage medium 1011 into the drive 1010. The program can also be received by the communication unit 1009 via a wired or wireless transmission medium and installed in the storage unit 1008. Alternatively, the program can be installed in advance in the ROM 1002 or the storage unit 1008.

[0187] The program executed by the computer may be a program that processes in chronological order according to the order described in this specification, or may be a program that processes in parallel or at the required timing, such as when called.

[0188] 16 realizes the functions of the control unit 41 in FIG.

[0189] In this specification, a system refers to a collection of multiple components (devices, modules (components), etc.), regardless of whether all of the components are contained in the same housing. Therefore, multiple devices housed in separate housings and connected via a network, and a single device with multiple modules housed in a single housing, are both systems.

[0190] Furthermore, the embodiments of the present disclosure are not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present disclosure.

[0191] For example, the present disclosure can be configured as a cloud computing system in which a single function is shared and processed collaboratively by multiple devices via a network.

[0192] Furthermore, each step described in the above flowchart can be executed by one device, or can be shared and executed by a plurality of devices.

[0193] Furthermore, when one step includes multiple processes, the multiple processes included in that one step can be executed by one device or can be shared and executed by multiple devices.

[0194] The present disclosure may also be configured as follows. <1> A program causing a computer to function as an image processing unit that extracts texture information from a captured image of the surface of a printed material captured at a predetermined resolution or higher, and a determination unit that determines whether the printed material has been printed using a predetermined printing method based on the texture information. <2> The predetermined printing method is a printing method using a reversible thermosensitive recording medium. <3> The program described in <1>, in which the image processing unit extracts the texture information by applying adaptive filtering to the captured image for each region and subtracting the result from the captured image. <4> The program described in <3>, in which the image processing unit extracts the texture information by applying adaptive filtering to the captured image for each region and subtracting the result from the captured image, and further normalizing each pixel by the variation of surrounding pixels. <5> The program described in <1>, further comprising a frequency analysis unit that performs frequency analysis of the texture information, and the determination unit determines whether the printed material has been printed using a predetermined printing method based on the frequency analysis result of the frequency analysis unit. <6> The program according to <5>, wherein the frequency analysis unit performs frequency analysis by applying a two-dimensional Fourier transform to the texture information and converting the two-dimensional Fourier transform result of the texture information into a power spectrum image consisting of a power distribution expressed by frequencies and angles at which the frequencies are detected. <7> The program according to <6>, wherein the frequency analysis unit outputs an image obtained by polar coordinate transformation of the power spectrum image as the frequency analysis result. <8> The program according to <7>, wherein the image obtained by polar coordinate transformation of the power spectrum image is an image in which the frequency axis and the angle axis in the two-dimensional Fourier transform result of the texture information are orthogonal. <9> The program according to <8>, further including a score calculation unit that calculates a score representing that the printed matter is printed using a predetermined printing method based on the frequency analysis result, and the determination unit determines whether the printed matter is printed using the predetermined printing method based on the score.<10> The program according to <9>, wherein the score calculation unit calculates the score based on an image obtained by polar coordinate transformation of the power spectrum image, which is the frequency analysis result, by masking frequency bands higher than a predetermined value and frequency bands lower than a predetermined value and masking an angle range of 180° to 360° in the image obtained by polar coordinate transformation of the power spectrum image, which is the frequency analysis result. <11> The program according to <10>, wherein the score calculation unit calculates, as a principal angle, an angle at which the sum of powers of all the frequencies for each angle is maximum in the image obtained by polar coordinate transformation of the power spectrum image, which is the frequency analysis result, and calculates the score for the principal angle. <12> The program described in <10>, wherein the score calculation unit calculates the frequency with the greatest power at the principal angle in an image obtained by polar coordinate transformation of the power spectrum image, which is the result of the frequency analysis, as the principal frequency, and calculates the score as the ratio of the power of the principal frequency at the principal angle to the sum of the power of all angles. <13> The program described in <6>, wherein the frequency analysis unit indicates that the captured image may have been out of focus when captured if the power spectrum in the power spectrum image is lower than a predetermined value. <14> The program described in <1>, further comprising an image extraction unit that extracts a recognition target region from the captured image, and the image processing unit extracts the texture information from the recognition target region extracted from the captured image. <15> The program described in <14>, wherein the recognition target region is a part of a living body included in the printed material. <16> The program described in <1>, wherein the predetermined resolution is 10 pixels / mm or more. <17> An information processing method including: image processing for extracting texture information from an image of a surface of a printed matter captured at a predetermined resolution or higher; and determination processing for determining whether the printed matter has been printed using a predetermined printing method based on the texture information.<18> The information processing method according to <17>, further including a frequency analysis process that performs frequency analysis by applying a two-dimensional Fourier transform to the texture information and converting the result of the two-dimensional Fourier transform of the texture information into a power spectrum image consisting of a power distribution expressed by frequencies and angles at which the frequencies are detected, wherein the determination process determines whether the printed matter has been printed using a predetermined printing method based on the power spectrum image that is the frequency analysis result of the frequency analysis process. <19> The information processing method according to <18>, wherein the frequency analysis process indicates that the captured image may have been out of focus if the power spectrum in the power spectrum image is too low. <20> An information processing system comprising: an image processing unit that extracts texture information from an image of the surface of a printed matter captured at a predetermined resolution or higher; and a determination unit that determines whether the printed matter has been printed using a predetermined printing method based on the texture information.

[0195] 31, 31', 31'' Information processing device, 50, 50', 50'' Camera, 51 Image extraction unit, 52 Preprocessing unit, 53 Frequency information extraction unit, 54 Score calculation unit, 55 Authenticity determination unit, 101, 101'' Information processing system

Claims

1. A program that causes a computer to function as an image processing unit that extracts texture information from an image of the surface of a printed matter captured at a specified resolution or higher, and a judgment unit that judges whether or not the printed matter has been printed using a specified printing method based on the texture information.

2. The program according to claim 1, wherein the predetermined printing method is a printing method using a reversible thermosensitive recording medium.

3. The program according to claim 1, wherein the image processing section extracts the texture information by applying adaptive filtering to the captured image for each region and subtracting the result from the captured image.

4. The program according to claim 3, wherein the image processing unit applies adaptive filtering to the captured image for each region, subtracts the result from the captured image, and further normalizes each pixel by the variation of surrounding pixels to extract the texture information.

5. The program according to claim 1, further comprising a frequency analysis unit that performs frequency analysis of the texture information, and the determination unit determines whether or not the printed matter has been printed using a predetermined printing method based on the results of the frequency analysis by the frequency analysis unit.

6. The program according to claim 5, wherein the frequency analysis unit performs frequency analysis by applying a two-dimensional Fourier transform to the texture information and converting the two-dimensional Fourier transform result of the texture information into a power spectrum image consisting of a distribution of power represented by frequency and the angle at which the frequency is detected.

7. The program according to claim 6, wherein the frequency analysis section outputs an image obtained by polar coordinate transformation of the power spectrum image as a result of the frequency analysis.

8. The program according to claim 7, wherein the image obtained by polar coordinate transformation of the power spectrum image is an image in which the frequency axis and the angle axis in the two-dimensional Fourier transform result of the texture information are orthogonal to each other.

9. The program according to claim 8, further comprising a score calculation unit that calculates a score representing that the printed matter has been printed using a predetermined printing method based on the frequency analysis results, and the judgment unit judges whether or not the printed matter has been printed using the predetermined printing method based on the score.

10. The program according to claim 9, wherein the score calculation unit calculates the score based on an image obtained by polar coordinate transformation of the power spectrum image, which is the result of the frequency analysis, by masking frequency bands higher than a predetermined value and frequency bands lower than a predetermined value and by masking an angle range of 180° to 360°.

11. The program according to claim 10, wherein the score calculation unit calculates the angle at which the sum of the power of all the frequencies for each angle is maximum in an image obtained by polar coordinate transformation of the power spectrum image, which is the result of the frequency analysis, as the principal angle, and calculates the score at the principal angle.

12. The program according to claim 10, wherein the score calculation unit calculates the frequency with the maximum power at the principal angle in the image obtained by polar coordinate transformation of the power spectrum image, which is the result of the frequency analysis, as the principal frequency, and calculates the ratio of the power of the principal frequency at the principal angle to the sum of the power of all angles as the score.

13. The program according to claim 6, wherein the frequency analysis unit indicates that the captured image may have been out of focus when captured if the power spectrum in the power spectrum image is lower than a predetermined value.

14. The program according to claim 1, further comprising an image extraction unit that extracts a recognition target area from the captured image, wherein the image processing unit extracts the texture information from the recognition target area extracted from the captured image.

15. The program according to claim 14, wherein the recognition target area is a part of a living body contained in the printed matter.

16. The program according to claim 1, wherein the predetermined resolution is 10 pixels / mm or more.

17. An information processing method including: image processing for extracting texture information from an image of the surface of a printed matter captured at a predetermined resolution or higher; and a determination process for determining whether or not the printed matter has been printed using a predetermined printing method based on the texture information.

18. An information processing method according to claim 17, further comprising a frequency analysis process for performing frequency analysis by applying a two-dimensional Fourier transform to the texture information and converting the result of the two-dimensional Fourier transform of the texture information into a power spectrum image consisting of a distribution of power represented by frequency and the angle at which the frequency is detected, and wherein the determination process determines whether or not the printed matter has been printed using a predetermined printing method based on the power spectrum image that is the frequency analysis result of the frequency analysis process.

19. The information processing method according to claim 18, wherein the frequency analysis process indicates that the captured image may have been out of focus when captured if the power spectrum in the power spectrum image is too low.

20. An information processing system comprising: an image processing unit that extracts texture information from an image of the surface of a printed material captured at a predetermined resolution or higher; and a judgment unit that judges whether the printed material has been printed using a predetermined printing method based on the texture information.

Images