Image matching device, image matching method, and program

Abstract

An image matching device 100 according to the present disclosure comprises: an acquisition unit 121 that acquires a hologram to be matched, the hologram being obtained by photographing, through holography, an object to be matched; and a matching unit 122 that matches the hologram to be matched with a registered hologram, which is a hologram that was registered in advance.

Description

Image matching device, image matching method, and program

[0001] The present disclosure relates to an image matching device, an image matching method, and a program.

[0002] For the purpose of individual identification, a so-called fingerprint of an object, which is a random pattern unique to the individual and appearing on the surface of the object, is extracted from an image obtained by photographing the object. For example, Patent Literature 1 describes a method of irradiating a laser onto the surface of an object to obtain a speckle image, and then performing individual identification by collating the speckle image.

[0003] JP 2011-233136 A

[0004] However, when performing image matching using speckle images as described in the above-mentioned Patent Document 1, the captured image may vary due to variations in the position of the object, resulting in a problem of reduced matching accuracy using images of the object.

[0005] Therefore, an object of the present disclosure is to solve the above-mentioned problem of reduced matching accuracy using images of objects.

[0006] An image matching device according to one aspect of the present disclosure includes: an acquisition unit that acquires a matching target hologram, which is a hologram obtained by holographically capturing an object to be matched; and a matching unit that matches the matching target hologram with a registered hologram, which is a hologram registered in advance. An image matching method according to one aspect of the present disclosure includes: acquiring a matching target hologram, which is a hologram obtained by holographically capturing an object to be matched, and matching the matching target hologram with a registered hologram, which is a hologram registered in advance. A program according to one aspect of the present disclosure includes: causing a computer to execute processing to acquire a matching target hologram, which is a hologram obtained by holographically capturing an object to be matched, and matching the matching target hologram with a registered hologram, which is a hologram registered in advance.

[0007] With the above-described configuration, the present disclosure can improve the accuracy of individual identification using an image of an object.

[0008] FIG. 1 is a block diagram showing the overall configuration of an image matching system according to the present disclosure; FIG. 2 is a block diagram showing the configuration of an image matching device according to the present disclosure; FIG. 3 is a diagram showing the state of processing by an image matching device according to the present disclosure; FIG. 4 is a flowchart showing the processing operation of an image matching device according to the present disclosure; FIG. 5 is a block diagram showing the hardware configuration of an image matching device according to the present disclosure; FIG. 6 is a block diagram showing the configuration of an image matching device according to the present disclosure.

[0009] First Embodiment A first embodiment of the present disclosure will be described with reference to the drawings. Note that the drawings may be relevant to any embodiment.

[0010] [Configuration] The image matching system in this embodiment uses a captured image of the object T to be matched to perform individual identification of the object T. For example, the object T is a pharmaceutical tablet itself, the sheet on which the tablet is packaged, a credit card, etc. In this embodiment, the image matching system extracts a Fingerprint of the Object, which is a random pattern unique to the object T that appears on the surface of the object T, from the captured image of the object T, and performs individual identification by checking whether the Fingerprint of the Object matches a registered image.

[0011] In this embodiment, a hologram image captured by holography is used as the image of the object T used for image matching. Therefore, the image matching system is configured to capture a hologram image by holography. Specifically, the image matching system includes a holographic optical system including a laser device L that irradiates the object T with laser light L0, a beam splitter BS that splits the laser light L0 into a reference light L2, and an image sensor 20 that detects a hologram generated by interference between the object light L1, which is light reflected by the object T from the laser light L0 irradiated on the object T, and the reference light L2. As a result, a hologram captured by the holographic optical system is configured as data recording interference fringes containing light intensity and phase information. Note that, although an off-axis holographic optical system is used in this embodiment, holograms may also be captured using holographic optical systems of other configurations, such as an in-line type, phase-shift type, or parallel phase-shift type.

[0012] The image matching system also includes an image matching device 10 that acquires a hologram image detected by an image sensor 20 and performs image matching. Since the above-mentioned holographic optical system is known as a general holography technology, a detailed description thereof will be omitted, and the following description will mainly focus on the image matching device.

[0013] Image matching device 10 is composed of one or more information processing devices each including a calculation device and a storage device. As shown in FIG. 2, image matching device 10 includes an acquisition unit 11, an extraction unit 12, and a matching unit 13. The functions of acquisition unit 11, extraction unit 12, and matching unit 13 can be realized by the calculation device executing a program for realizing each function stored in the storage device. Image matching device 10 also includes a registered image storage unit 16 composed of a storage device. Each component will be described in detail below.

[0014] The registered image storage unit 16 stores registered holograms, which are holograms that have been registered in advance. A registered hologram is a hologram obtained by holographically capturing the surface of an existing object, and is data that records interference fringes containing light intensity and phase information detected by the holographic optical system described above. For example, an object with a registered hologram may be an object that has been approved as a genuine product or an object with a registered serial number.

[0015] The acquisition unit 11 acquires a matching target hologram, which is a hologram obtained by holographically capturing an image of an object T, from the above-described holographic optical system, i.e., the image sensor 20. Here, the object T from which the matching target hologram is captured is an object T to be matched with an object in which the above-described registered hologram is registered, such as an object T whose authenticity and serial number are unknown. The matching target hologram is configured as data recording interference fringes including light intensity and phase information detected by the image sensor 20.

[0016] The extraction unit 12 extracts an object light image from a hologram to be compared and generates a match image including the object light image. Specifically, the extraction unit 12 first performs a two-dimensional Fourier transform on the hologram to be compared to obtain a power spectrum image. The extraction unit 12 then extracts the object light image from the power spectrum image. For example, the extraction unit 12 selects pixels with high brightness values ​​from the power spectrum image to obtain a zeroth-order diffracted light image g0, an object light image g1 consisting of +1st-order diffracted light, and a conjugate light image g2 consisting of −1st-order diffracted light, as shown by reference symbol G1 in FIG. 3 . However, by performing a preset object light extraction calculation, only the object light image g1 is extracted, as shown by reference symbol G2 in FIG. 3 . The extraction unit 12 then generates a match image G3 by shifting the extracted object light image g1 to its center position, as shown by reference symbol G3 in FIG. 3 .

[0017] Here, the method for generating the above-mentioned match image G3 by the extraction unit 12 will be described in more detail. First, as shown in the left diagram of FIG. 4 , a power spectrum is obtained by Fourier transforming a match target hologram obtained by detecting the object light L1 and reference light L2 reflected from an object T irradiated with laser light using the image sensor 20. This allows for the acquisition of a zeroth-order diffracted light image g0, an object light image g1 consisting of +1st-order diffracted light, and a conjugate light image g2 consisting of −1st-order diffracted light. The right diagram of FIG. 4 shows the state when the tilt angle of the object T is changed by θ compared to the left diagram. In this case, as shown in the power spectrum obtained by Fourier transforming the match target hologram, the position of the object light image g1 consisting of +1st-order diffracted light moves in accordance with the tilt angle θ of the object T relative to the position before the tilt angle is changed (the power spectrum indicated by the dotted line). At the same time, the position of the conjugate light image g2 consisting of −1st-order diffracted light also moves symmetrically about the image center by the same distance as the object light image g relative to the position before the tilt angle is changed (the power spectrum indicated by the dotted line). In this way, the amount of movement of the object light image g1 and the distance from the image center can be determined according to the value of the tilt angle. Therefore, by setting an allowable range that represents an area in which movement of the object light position in accordance with changes in the tilt angle is allowed with respect to the appearance position of the object light that is assumed in advance on the hologram to be compared, it becomes possible to extract only the object light image g1.

[0018] A specific example of the process performed by the extraction unit 12 to extract the object light image g1 and generate the match image G3 will be described below. First, the extraction unit 12 selects the pixel with the highest brightness value within the allowable range of the object light image set as described above from the hologram to be matched. Next, the extraction unit 12 sets a circular filter centered on the selected pixel and sets the pixel values ​​outside the filter to 0. The circular filter is set to a size large enough to accommodate the object light image g1 and not overlap with the zeroth-order diffracted light image g0, which is located at the center of the image when set within the allowable range of the object light image. Then, the extraction unit 12 shifts the pixel distribution within the circular filter so that the selected pixel is located at the center of the image. In this way, the extraction unit 12 can generate the match image G3 in which the object light image g1 is located at the center of the image, as shown in FIG. 3 . However, the extraction unit 12 is not necessarily limited to generating the match image G3 using the method described above and may generate the match image G3 using any method.

[0019] The extraction unit 12 also performs the same processing as described above on the registered hologram, extracting the object light image in the registered hologram and generating a match image in which the object light image is located at the center of the image. However, the match image of the registered hologram may be generated in advance and registered in the registered image storage unit 16.

[0020] As described above, the matching unit 13 matches the matching image G3 of the target hologram with the matching image of the registered hologram. For example, the matching unit 13 performs a logarithmic polar coordinate transformation based on the power spectrum image for each of the matching image G3 of the target hologram and the matching image of the registered hologram to calculate polar coordinate images, i.e., Fourier-Mellin feature images. The matching unit 13 then performs a discrete Fourier transform on the Fourier-Mellin feature images corresponding to each of the matching image G3 of the target hologram and the matching image of the registered hologram to extract phase components, i.e., Fourier-Mellin phase features, and performs a matching determination by comparing these. For example, the matching unit 13 calculates a matching score by performing a predetermined correlation calculation of the Fourier-Mellin phase features corresponding to each of the matching image G3 of the target hologram and the matching image of the registered hologram, and performs a matching determination based on the matching score. As an example, if the matching score is equal to or greater than a threshold, it is determined that the matching image G3 of the hologram to be matched and the matching image of the registered hologram match, and if the matching score is less than the threshold, it is determined that they do not match. Note that the matching method using the Fourier-Mellin feature image by the above-mentioned matching unit 13 is known, so a detailed description thereof will be omitted.

[0021] Furthermore, during the above-described matching process, the matching unit 13 may generate multiple match images G3 for the matching target hologram reconstructed at multiple focal lengths, and match each of the multiple match images G3 with the match image of the registered hologram. For example, the matching unit 13 may generate match images G3 including object images reconstructed at multiple focal lengths using intensity information and phase information from the power spectrum of the match image G3 including the object light image g1 extracted as described above. The matching unit 13 may then perform a final matching determination for the matching target hologram based on multiple matching results obtained by comparing each of the multiple match images G3 with the match image of the registered hologram. As an example, the matching unit 13 may identify one match result that satisfies a predetermined criterion, such as the highest matching score, and perform the matching determination by comparing the value of the identified match result with a threshold. Note that the matching unit 13 may also perform a final matching determination based on multiple matching results, such as by comparing a value such as the average or mode of the matching scores for multiple match images G3 reconstructed at different focal lengths with a threshold. In this case, the matching unit 13 may use a plurality of matching images generated by reconstructing the registered hologram at each of a plurality of focal lengths in the same manner as described above.

[0022] The above-described matching process by the matching unit 13 is merely an example, and the matching between the matching target hologram and the registered hologram may be performed by any method. For example, the matching between the matching image G3 of the matching target hologram and the matching image of the registered hologram may be performed by normalized cross-correlation or feature point-based matching.

[0023] [Operation] Next, a description will be given of the operation of the above-described image matching device 10. At this time, a registered hologram is stored in advance in image matching device 10. Note that, for the registered hologram, a matching image for matching made up of an object light image may be generated and stored.

[0024] First, image matching device 10 acquires a target hologram obtained by capturing an image of object T to be matched using a holographic optical system such as that shown in FIG. 1 (step S1 in FIG. 5 ). Image matching device 10 then extracts an object light image from the acquired target hologram (step S2 in FIG. 5 ). Specifically, image matching device 10 performs a two-dimensional Fourier transform on the target hologram, such as that shown by reference character G1 in FIG. 3 , to acquire a power spectrum image, and extracts object light image g1 from the power spectrum image, as shown by reference character G2 in FIG. 3 . Image matching device 10 then generates match image G3 by shifting the extracted object light image g1 to its center position, as shown by reference character G3 in FIG. 3 (step S3 in FIG. 5 ). As described above, image matching device 10 can extract only object light image g1 by setting an allowable range within which an object light image may appear on the target hologram, depending on the tilt angle of object T.

[0025] Thereafter, image matching device 10 matches matching image G3 of the target hologram with the matching image of the registered hologram (step S4 in FIG. 5). At this time, image matching device 10 may generate multiple matching images G3 by reconstructing the target hologram at multiple focal lengths, and match each of the multiple matching images G3 with the matching image of the registered hologram.

[0026] As described above, the image matching device 10 of this embodiment uses a hologram as the image to be matched, thereby suppressing the effects of positional fluctuations of the object and obtaining highly accurate matching results, thereby improving the accuracy of individual identification according to the matching results. In particular, the use of a hologram makes it possible to suppress the effects of fluctuations in the elevation angle of the object T and obtain more accurate matching results. Furthermore, the image matching device 10 of this embodiment can perform matching using images reproduced at multiple focal lengths using a hologram. As a result, it is possible to suppress the effects of positional fluctuations in the optical axis direction of the object T and obtain more accurate matching results.

[0027] Second Embodiment Next, a second embodiment of the present disclosure will be described with reference to the drawings. This embodiment shows an outline of the configuration of the image matching device described in the above embodiment. Note that Figures 6 and 7 are diagrams for explaining the configuration, and these drawings may be relevant to any of the embodiments.

[0028] First, the hardware configuration of image matching device 100 will be described with reference to Fig. 6. Image matching device 100 is configured as a general information processing device, and is equipped with the following hardware configuration, for example: CPU (Central Processing Unit) 101 (arithmetic unit) ROM (Read Only Memory) 102 (storage device) RAM (Random Access Memory) 103 (storage device) Programs 104 loaded into RAM 103 Storage device 105 for storing programs 104 Drive device 106 for reading and writing data from / to storage medium 110 outside the information processing device Communication interface 107 for connecting to communication network 111 outside the information processing device Input / output interface 108 for inputting and outputting data Bus 109 for connecting each component

[0029] 6 shows an example of the hardware configuration of the information processing device that is the image matching device 100, and the hardware configuration of the information processing device is not limited to the above-described case. For example, the information processing device may be configured with a part of the above-described configuration, such as not including the drive device 106. Furthermore, the information processing device may use a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), an MPU (Micro Processing Unit), an FPU (Floating point number Processing Unit), a PPU (Physics Processing Unit), a TPU (Tensor Processing Unit), a quantum processor, a microcontroller, or a combination thereof, instead of the above-described CPU.

[0030] 7 can be constructed and equipped with an acquisition unit 121 and a matching unit 122 shown in FIG. 7 by having CPU 101 acquire and execute program group 104. Program group 104 is stored in advance in storage device 105 or ROM 102, for example, and is loaded into RAM 103 and executed by CPU 101 as needed. Program group 104 may be supplied to CPU 101 via communication network 111, or may be stored in advance in storage medium 110, with drive device 106 reading out the programs and supplying them to CPU 101. However, the acquisition unit 121 and matching unit 122 described above may be constructed using dedicated electronic circuits for realizing such means.

[0031] The acquisition unit 121 acquires a hologram to be verified, which is a hologram obtained by photographing an object to be verified by holography. The verification unit 122 then verifies the hologram to be verified with a registered hologram, which is a hologram that has been registered in advance.

[0032] With the above-described configuration, the present disclosure uses a hologram as the image to be compared, thereby suppressing the effects of fluctuations in the object's position and obtaining highly accurate comparison results. In particular, by using a hologram, the effects of fluctuations in the object's tilt angle and position in the optical axis direction can be suppressed, resulting in more accurate comparison results.

[0033] In addition, at least one of the functions of the above-mentioned acquisition unit 121 and matching unit 122 may be executed by an information processing device installed and connected anywhere on the network, that is, they may be executed by so-called cloud computing.

[0034] The above-described program can be stored and supplied to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-RWs, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program can also be supplied to a computer by various types of transitory computer-readable media. Examples of transitory computer-readable media include electrical signals, optical signals, and electromagnetic waves. The transitory computer-readable media can be supplied to a computer via wired communication paths such as electric wires and optical fibers, or via wireless communication paths.

[0035] Although the present disclosure has been described above with reference to the above-described embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each of the above-described embodiments can be combined with other embodiments as appropriate.

[0036] <Supplementary Notes> Some or all of the above embodiments can also be described as in the following supplementary notes. Below, an outline of the configurations of an image matching device, an image matching method, and a program according to the present disclosure will be described. However, the present disclosure is not limited to the following configurations. (Supplementary Note 1) An image matching device comprising: an acquisition unit that acquires a target hologram, which is a hologram obtained by holographically capturing an image of an object to be matched; and a matching unit that matches the target hologram with a registered hologram, which is a hologram that has been registered in advance. (Supplementary Note 2) The image matching device according to Supplementary Note 1, further comprising: an extraction unit that extracts an object light image from the target hologram and generates a match image including the object light image, and the matching unit matches the match image with the registered hologram. (Supplementary Note 3) The image matching device according to Supplementary Note 2, wherein the extraction unit moves the object light image extracted from the target hologram to a center position of the match image to generate the match image. (Supplementary Note 4) The image matching device according to Supplementary Note 2, wherein the extraction unit extracts the object light image based on a power spectrum image of the target hologram for matching. (Supplementary Note 5) The image matching device according to Supplementary Note 4, wherein the extraction unit extracts the object light image based on a power spectrum image of the target hologram for matching from within an area on the target hologram for matching set corresponding to a range of tilt angles of the object allowed during imaging. (Supplementary Note 6) The image matching device according to Supplementary Note 2, wherein the matching unit generates a plurality of match images including images obtained by reconstructing the object light image at a plurality of focal lengths, and makes a matching determination based on a result of matching each of the plurality of match images with the registered hologram. (Supplementary Note 7) The image matching device according to Supplementary Note 6, wherein the matching unit makes a matching determination based on one of the results of matching each of the plurality of match images with the registered hologram that satisfies a preset criterion.(Supplementary Note 8) An image matching method comprising: acquiring a comparison target hologram, which is a hologram obtained by photographing an object to be compared by holography; and matching the comparison target hologram with a registered hologram, which is a hologram that has been registered in advance. (Supplementary Note 9) The image matching method according to Supplementary Note 8, comprising: extracting an object light image from the acquired comparison target hologram to generate a comparison image including the object light image; and matching the comparison image with the registered hologram. (Supplementary Note 10) The image matching method according to Supplementary Note 9, comprising: moving the object light image extracted from the acquired comparison target hologram to a center position of the comparison image to generate the comparison image. (Supplementary Note 11) The image matching method according to Supplementary Note 9, comprising: generating a plurality of comparison images including images obtained by reconstructing the object light image at each of a plurality of focal lengths; and making a matching determination based on the results of comparing each of the plurality of comparison images with the registered hologram. (Supplementary Note 12) An image matching device comprising: an acquisition unit that acquires a comparison target hologram, which is a hologram obtained by holographically capturing an image of an object to be matched; and a matching unit that matches the comparison target hologram with a registration hologram, which is a hologram that has been registered in advance. (Supplementary Note 13) A computer-readable storage medium that stores a program that causes a computer to execute the following processes: acquire a comparison target hologram, which is a hologram obtained by holographically capturing an image of an object to be matched, and match the comparison target hologram with a registration hologram, which is a hologram that has been registered in advance. (Supplementary Note 14) A computer-readable storage medium that stores the program according to Supplementary Note 13, comprising: extracting an object light image from the acquired comparison target hologram to generate a comparison image including the object light image, and matching the comparison image with the registration hologram.

[0037] REFERENCE SIGNS LIST 10 Image matching device 11 Acquisition unit 12 Extraction unit 13 Matching unit 16 Registered image storage unit 20 Image sensor L Laser device BS Beam splitter T Object 100 Image matching device 101 CPU 102 ROM 103 RAM 104 Program group 105 Storage device 106 Drive device 107 Communication interface 108 Input / output interface 109 Bus 110 Storage medium 111 Communication network 121 Acquisition unit 122 Matching unit

Claims

1. An image matching device comprising: an acquisition unit that acquires a matching target hologram which is a hologram obtained by photographing an object to be matched by holography; and a matching unit that matches the matching target hologram with a registered hologram which is a hologram registered in advance.

2. The image matching device according to claim 1, further comprising: an extraction unit that extracts an object light image from the matching target hologram and generates a matching image including the object light image; wherein the matching unit matches the matching image with the registered hologram.

3. The image matching device according to claim 2, wherein the extraction unit generates the matching image by moving the object light image extracted from the matching target hologram to the center position of the matching image.

4. The image matching device according to claim 2, wherein the extraction unit extracts the object light image based on a power spectrum image of the matching target hologram.

5. The image matching device according to claim 4, wherein the extraction unit extracts the object light image based on the power spectrum image of the matching target hologram from within a region on the matching target hologram set corresponding to a range of sway angles of the object allowed during photographing.

6. The image matching device according to claim 2, wherein the matching unit generates a plurality of the matching images each including an image reproduced from the object light image at each of a plurality of focal lengths, and performs a matching determination based on results of matching each of the plurality of the matching images with the registered hologram.

7. The image matching device according to claim 6, wherein the matching unit performs a matching determination based on one of the matching results of each of the plurality of the matching images with the registered hologram that satisfies a preset criterion.

8. An image matching method comprising: acquiring a matching target hologram which is a hologram obtained by photographing an object to be matched by holography; and matching the matching target hologram with a registered hologram which is a hologram registered in advance.

9. The image matching method according to claim 8, further comprising: extracting an object light image from the acquired matching target hologram and generating a matching image including the object light image; and matching the matching image with the registered hologram.

10. The image matching method according to claim 9, wherein the object light image extracted from the acquired hologram to be matched is moved to the center position of the matching image to generate the matching image.

11. The image matching method according to claim 9, wherein a plurality of the matching images each including an image reproduced from the object light image at each of a plurality of focal lengths are generated, and a matching determination is made based on results of matching each of the plurality of the matching images with the registered hologram.

12. An image matching apparatus comprising: an acquisition unit that acquires a hologram to be matched, which is a hologram obtained by photographing an object to be matched by holography; and a matching unit that matches the hologram to be matched with a registered hologram that is a hologram registered in advance.

13. A computer-readable storage medium storing a program for causing a computer to execute a process of acquiring a hologram to be matched, which is a hologram obtained by photographing an object to be matched by holography, and matching the hologram to be matched with a registered hologram that is a hologram registered in advance.

14. A computer-readable storage medium storing a program for causing a computer to execute a process of extracting an object light image from the acquired hologram to be matched to generate a matching image including the object light image, and matching the matching image with the registered hologram.

Images