Biological data correction device, biological data correction method, and biological data correction system

Abstract

Provided is a biological data correction device comprising: an acquisition unit that acquires a fingerprint image of a fingerprint captured in a non-contact state; a detection unit that detects a first center position of the fingerprint in the fingerprint image; a calculation unit that calculates a plurality of correction control points for correcting the first center position of the fingerprint to a second center position in the fingerprint image on the basis of the first center position of the fingerprint with respect to a field of view of the fingerprint image; and a correction unit that outputs a corrected image obtained by correcting coordinates of pixels in the fingerprint image on the basis of the plurality of correction control points.

Description

Biological data correction device, biological data correction method, and biological data correction system

[0001] The present disclosure relates to a biological data correction device, a biological data correction method, and a biological data correction system.

[0002] In Patent Document 1, in a fingerprint image shown by fingerprint image data in which a person's fingerprint is captured, or a gradation value indicating the gradation of each pixel constituting the fingerprint image, or a difference obtained from the gradation value of each pixel and the gradation value of surrounding pixels. In the extraction image created based on the information, regarding the fingerprint image or the extraction image as a two-dimensional plane, on the two-dimensional plane, the first approximation that approximates the set of points on the coordinates corresponding to the pixels where the fingerprint is supposed to be projected to an approximate straight line. The difference in inclination between the first approximation straight line, which is the approximated approximate straight line, and a reference line, which is a line predetermined for the fingerprint image, is calculated as a rotation angle, which is the angle by which the fingerprint area, which is the area where the fingerprint exists, is inclined with respect to the reference line. A fingerprint image rotation angle calculation device is disclosed.

[0003] Japanese Patent Application Laid-Open No. 2018-217090

[0004] The present disclosure has been devised in view of the above-described conventional circumstances, and an object thereof is to provide a biological data correction device, a biological data correction method, and a biological data correction system that correct the angle of a biological part included in a biological image acquired in a free space.

[0005] The present disclosure provides a biological data correction device including: an acquisition unit that acquires a fingerprint image of a fingerprint imaged in a non-contact state; a detection unit that detects a first center position of the fingerprint shown in the fingerprint image; a calculation unit that calculates a plurality of correction control points for correcting the first center position of the fingerprint to a second center position of the fingerprint image based on the first center position of the fingerprint with respect to the picture angle of the fingerprint image; and a correction unit that outputs a correction image obtained by correcting the coordinates of pixels on the fingerprint image based on the plurality of correction control points.

[0006] Furthermore, this disclosure provides a biometric data correction method performed by at least one processor, which includes acquiring a fingerprint image of a fingerprint captured in a non-contact state, detecting a first central position of the fingerprint as depicted in the fingerprint image, calculating a plurality of correction control points that correct the first central position of the fingerprint to a second central position of the fingerprint image based on the first central position of the fingerprint with respect to the field of view of the fingerprint image, and outputting a corrected image in which the coordinates of pixels on the fingerprint image have been corrected based on the plurality of correction control points.

[0007] Furthermore, this disclosure provides a biometric data correction system comprising: an imaging device for capturing a fingerprint of a person to be authenticated in a non-contact state; and a correction device capable of communicating with the imaging device, wherein the imaging device transmits an image of at least one fingertip to the correction device; the correction device extracts a region including the first joint of the fingertip from the image to generate a fingerprint image of the fingerprint; detects a first central position of the fingerprint in the fingerprint image; calculates a plurality of correction control points that correct the first central position of the fingerprint to a second central position of the fingerprint image based on the first central position of the fingerprint with respect to the field of view of the fingerprint image; and outputs a corrected image in which the coordinates of pixels on the fingerprint image have been corrected based on the plurality of correction control points.

[0008] According to this disclosure, the angles of biological parts included in biological images acquired in free space can be corrected.

[0009] Figure showing an example of the overall configuration of the biometric data correction system according to Embodiment 1. Block diagram showing an example of the internal configuration of the biometric data correction system according to Embodiment 1. Flowchart showing an example of the operation procedure of the correction device in this disclosure. Figure showing an example of correction control points before and after correction. Figure showing an example of fingerprint images and authentication fingerprint images before and after correction. Figure showing an example of the overall configuration of the biometric data correction system according to Embodiment 2. Block diagram showing an example of the internal configuration of the biometric data correction system according to Embodiment 2.

[0010] (Background to this disclosure) Generally, biometric authentication using fingerprints authenticates a registered person and a person requesting authentication by matching the characteristic points that indicate the individuality of the fingerprint, which are contained in the fingerprint data registered at the time of registration and the fingerprint data acquired for authentication. Therefore, a problem with biometric authentication is that if there is a difference in the central position between the fingerprint registered at the time of registration and the fingerprint acquired for authentication, the characteristic points of each fingerprint will also be shifted accordingly, resulting in a decrease in authentication accuracy.

[0011] Here, there are two methods for obtaining fingerprints used in biometric authentication: one involves acquiring fingerprints from a user's body part while that body part is in contact with a predetermined surface, and the other involves acquiring fingerprints from a user's fingers in a non-contact state, where they are held in free space. Fingerprints acquired in a contact state are pressed against the predetermined surface, which can lead to discrepancies in the orientation of the fingerprint relative to that surface. In particular, fingerprints acquired in a non-contact state are held in free space, which presents a challenge because discrepancies are likely to occur not only in the orientation of the fingerprint but also in its angle (position).

[0012] The fingerprint image rotation angle calculation device described above corrects the tilt (orientation) of the fingerprint in the fingerprint image by rotating the fingerprint image or extraction image based on the rotation angle of the fingerprint area captured in the fingerprint image in a non-contact manner. However, as mentioned above, fingerprints acquired in a non-contact manner are prone to discrepancies in both the orientation and angle (orientation) of the fingerprint. Therefore, there has been a demand for a method to correct discrepancies in the angle (orientation) of fingerprints in biometric authentication.

[0013] Hereinafter, with reference to the drawings as appropriate, each embodiment that specifically discloses the configuration and operation of the biometric data correction device, biometric data correction method, and biometric data correction system relating to this disclosure will be described in detail. However, unnecessarily detailed explanations may be omitted. For example, detailed explanations of already well-known matters or redundant explanations of substantially identical configurations may be omitted. This is to avoid the following explanation becoming unnecessarily redundant and to facilitate understanding by those skilled in the art. The attached drawings and the following explanation are provided to enable those skilled in the art to fully understand this disclosure and are not intended to limit the subject matter described in the claims.

[0014] (Embodiment 1) The biometric data correction system 100 according to Embodiment 1 will be described with reference to Figures 1 and 2, respectively. Figure 1 is a diagram showing an example of the overall configuration of the biometric data correction system 100 according to Embodiment 1. Figure 2 is a block diagram showing an example of the internal configuration of the biometric data correction system 100 according to Embodiment 1.

[0015] The biometric data correction system 100 in Embodiment 1 generates a fingerprint image used for fingerprint authentication by correcting an image captured by a fingerprint acquisition device P1 with a correction device P2. The biometric data correction system 100 includes, as an example, a fingerprint acquisition device P1 and a correction device P2. The configuration of the biometric data correction system 100 shown in Figures 1 and 2 is an example and is not limited thereto. Other configurations will be described later. The biometric data correction system 100 in this disclosure can not only generate an authentication fingerprint image used for fingerprint authentication, but also register a fingerprint image used for fingerprint authentication, or perform fingerprint authentication using an authentication fingerprint image.

[0016] The fingerprint acquisition device P1 is connected to the correction device P2 via wired or wireless communication to perform data transmission and reception. The fingerprint acquisition device P1 is a device that captures the user's fingerprint used for fingerprint authentication, and can be implemented, for example, by a camera, smartphone, or tablet terminal. The fingerprint acquisition device P1 includes a communication unit 10, a processor 11, and a memory 12.

[0017] The communication unit 10 is connected to the correction device P2 via wireless or wired communication. Wireless communication here refers to, for example, short-range wireless communication such as Bluetooth® or NFC®, or communication via a wireless Local Area Network (LAN) such as Wi-Fi®. The communication unit 10 transmits the captured image captured by the imaging unit 13 to the correction device P2. The communication unit 10 also outputs various data or information transmitted from the correction device P2 to the processor 11.

[0018] The processor 11 is configured using, for example, a Central Processing Unit (hereinafter referred to as "CPU"), a Field Programmable Gate Array (hereinafter referred to as "FPGA"), or a Graphics Processing Unit (hereinafter referred to as "GPU"), and works in cooperation with the memory 12 to perform various processes and controls. Specifically, the processor 11 refers to the programs and data held in the memory 12 and executes those programs to realize the functions of the fingerprint acquisition device P1.

[0019] Memory 12 includes, for example, Random Access Memory (hereinafter referred to as "RAM"), which serves as work memory used when executing each process of the processor 11, and Read Only Memory (hereinafter referred to as "ROM"), which stores programs and data that define the operation of the processor 11. Data or information generated or acquired by the processor 11 is temporarily stored in RAM. Programs that define the operation of the processor 11 are written in ROM.

[0020] The imaging unit 13 is implemented, for example, by an optical system including a lens and an image sensor, and captures fingerprints in a non-contact state with any object. The imaging unit 13 may capture one fingertip (fingerprint) in a single capture, or it may capture multiple fingertips (fingerprints) simultaneously in a single capture. The imaging unit 13 outputs the captured image to the processor 11.

[0021] The correction device P2 is connected to the fingerprint acquisition device P1 via wired or wireless communication and performs data transmission and reception. The correction device P2 acquires a fingerprint image by extracting the area containing the fingerprint from the captured image transmitted from the fingerprint acquisition device P1, corrects this fingerprint image, and generates an authentication fingerprint image for use in fingerprint authentication. The correction device P2 includes a communication unit 20, a processor 21, a memory 22, a fingerprint acquisition unit 23, a monitor 24, and a registered fingerprint database DB.

[0022] Note that the fingerprint acquisition unit 23, monitor 24, and registered fingerprint database DB are not mandatory components and may be omitted. For example, the fingerprint acquisition unit 23 performs the same function as the imaging unit 13 of the fingerprint acquisition device P1 when the fingerprint acquisition device P1 and the correction device P2 are configured as an integrated unit. For example, the monitor 24 may be implemented by an external monitor capable of communicating with the correction device P2. Also, for example, the registered fingerprint database DB may be configured separately from the correction device P2, and may be omitted if the correction device P2 does not perform fingerprint authentication.

[0023] The communication unit 20 is connected to the fingerprint acquisition device P1 via wireless or wired communication. The communication unit 20 transmits various data or information received from the processor 21 to the fingerprint acquisition device P1. The communication unit 20 also outputs various data (e.g., captured images) or information received from the fingerprint acquisition device P1 to the processor 21.

[0024] The processor 21 is configured using, for example, a CPU, FPGA, or GPU, and works in cooperation with the memory 22 to perform various processes and controls. Specifically, the processor 21 refers to the programs and data held in the memory 22 and executes those programs to realize the functions of each part. These parts include, for example, the correction unit 211, the registration unit 212, and the authentication unit 213. Note that the registration unit 212 and the authentication unit 213 are not mandatory and may be omitted.

[0025] The correction unit 211 analyzes the captured image taken by the fingerprint acquisition device P1 and detects each finger in the fingerprint image and the area in which the fingerprint of each finger is captured (hereinafter referred to as the "fingerprint area"). The fingerprint area is an area that includes at least the first joint of the fingertip. The correction unit 211 generates a fingerprint image IMG 11 for each finger by cutting out the detected fingerprint area. The correction unit 211 corrects the generated fingerprint image IMG 11 for each finger into an authentication fingerprint image IMG 12 suitable for fingerprint authentication. Specifically, the correction unit 211 generates an authentication fingerprint image IMG 12 in which the position and orientation of the fingerprints in the fingerprint image are corrected so that the fingerprint is located approximately in the center of the fingerprint image, in an image taken from the front of the fingerprint.

[0026] The registration unit 212 performs the registration (storage) of fingerprint images of users whose fingerprint images used for biometric authentication have not yet been registered. When the registration unit 212 performs the registration of a fingerprint image, it obtains the authentication fingerprint image corrected by the correction unit 211 and registers (stores) it in the registered fingerprint database DB as the registered fingerprint image used for biometric authentication.

[0027] The authentication unit 213 acquires the authentication fingerprint image IMG12 corrected by the correction unit 211. The authentication unit 213 compares the authentication fingerprint image IMG12 with at least one registered fingerprint image registered in the registered fingerprint database DB and performs fingerprint authentication. The authentication unit 213 outputs the authentication result to the monitor 24. Note that the destination for sending and outputting the authentication result is not limited to the example described above.

[0028] Memory 22 includes, for example, RAM as work memory used when executing each process of the processor 21, and ROM which stores programs and data that define the operation of the processor 21. Data or information generated or acquired by the processor 21 is temporarily stored in RAM. Programs that define the operation of the processor 21 are written in ROM. Memory 22 also stores various data, various software, and various programs for realizing functions such as the correction unit 211 or the authentication unit 213.

[0029] The fingerprint acquisition unit 23 is implemented, for example, by an optical system including a lens and an image sensor, and captures a fingerprint in a non-contact state with any object. The fingerprint acquisition unit 23 may capture one fingertip (fingerprint) in a single capture, or it may capture multiple fingertips (fingerprints) simultaneously in a single capture. The fingerprint acquisition unit 23 outputs the captured image to the processor 21 as an image containing a fingerprint to be used for authentication or registration.

[0030] The monitor 24 is configured using, for example, a Liquid Crystal Display (hereinafter referred to as "LCD") or an organic Electroluminescence (hereinafter referred to as "EL"). The monitor 24 outputs and displays various screens that notify the user of captured images taken by the fingerprint acquisition device P1 or the fingerprint acquisition unit 23, or the authentication results from the authentication unit 213.

[0031] The registered fingerprint database DB is a so-called storage system, configured using storage media such as flash memory, a Hard Disk Drive (HDD), or a Solid State Drive (SSD). The registered fingerprint database DB stores (registers) registered fingerprint images, which are used for biometric authentication and matched with authentication fingerprint images, and information about the user corresponding to these registered fingerprint images, linked together for each user. If multiple fingerprints are registered for a single user, the registered fingerprint database DB may further link and store information indicating whether the fingerprints in the registered fingerprint images belong to the left or right hand.

[0032] <Method for correcting fingerprint images> Next, a method for correcting fingerprint images will be described with reference to Figure 3. Figure 3 is a flowchart showing an example of the operation procedure of the correction device P2 in this disclosure. Figure 4 is a diagram showing an example of correction control points before and after correction. Figure 5 is a diagram showing an example of a fingerprint image IMG11 and an authentication fingerprint image IMG12. Note that the operation procedure shown in Figure 3 is a process that is executed for each fingerprint image from which the fingerprint of each finger has been extracted.

[0033] The processor 21 acquires the user's fingerprint image IMG 11 (St 11). The processor 21 detects the center position of the fingerprint in the fingerprint image IMG 11 (St 12).

[0034] Furthermore, known techniques may be used to detect the central position of a fingerprint. For example, it may be detected using a rule-based method based on manutsier features, or it may be detected using a deep learning model capable of detecting the central position of a fingerprint.

[0035] Here, the central position of the fingerprint in this disclosure will be explained. Fingerprint patterns include many types classified based on the characteristics of the pattern, such as loop-shaped patterns, arch-shaped patterns, whorl patterns, variant patterns, type A loop-shaped patterns, or type B loop-shaped patterns. Furthermore, fingerprints tend to have patterns depending on the type of finger (e.g., index finger, middle finger) depending on race or ethnicity. Therefore, the central position of the fingerprint in this disclosure is not limited to the literal central position of the fingerprint pattern, but may be defined for each type of fingerprint (pattern). In such cases, the processor 21 may perform pattern estimation based on the fingerprint pattern captured in the fingerprint image, and then detect the central position of the fingerprint defined based on the estimated pattern. Known techniques may be used for this pattern estimation method. Also, for example, if the biometric data correction system 100 is used in a place where the user's race, genes, or nationality can be obtained using an identification document such as a passport, such as a hospital, airport, or port, the processor 21 may perform pattern estimation using the obtained user's race, genes, or nationality.

[0036] The processor 21 calculates the confidence level (detection confidence level) of the detected center position of the fingerprint (St13). The processor 21 determines whether the calculated detection confidence level is above a threshold (St13). Note that known techniques may be used for the detection confidence level calculation process. Furthermore, the calculation and determination processes for the detection confidence level of the center position of the fingerprint (step St13) are not mandatory processes and may be omitted.

[0037] If the processor 21 determines that the calculated detection confidence level is below a threshold (St13, NO), it determines that it is difficult to generate an authentication fingerprint image suitable for fingerprint authentication through correction, and omits the fingerprint image correction process. The processor 21 performs fingerprint authentication by comparing the uncorrected fingerprint image IMG11 with the registered fingerprint image (St17).

[0038] On the other hand, if the processor 21 determines that the calculated detection confidence level is above a threshold (St13, YES), it starts correcting (converting) the fingerprint image IMG11 in order to generate an authentication fingerprint image IMG12, which is an image of the fingerprint taken from the front, in which the center position Pt11 of the fingerprint is captured at the approximate center position Pt12 of the field of view.

[0039] The processor 21 calculates control points (hereinafter referred to as "correction control points") that transform the pixels of the entire fingerprint image IMG 11. The processor 21 performs spline interpolation (specifically, Thin Plate Spline (TPS processing)) based on the center position Pt 11 of the fingerprint relative to both ends of the X-axis and Y-axis directions of the fingerprint image IMG 11 (the peripheral portion of the field of view of the fingerprint image IMG 11). The processor 21 calculates each of the multiple correction control points (St 14). The number of correction control points may be arbitrary.

[0040] For example, in the example shown in Figure 4, the processor 21 performs spline interpolation based on the position of the center position Pt11 of the fingerprint relative to both ends of the fingerprint image IMG11 in the X-axis and Y-axis directions (the field of view of the fingerprint image IMG11). By performing spline interpolation, the processor 21 calculates a total of 100 correction control points Cn01A to Cn100A, with the center position Pt11 of the fingerprint as the center, and five points each extending from the center position Pt11 in the X-axis and Y-axis directions.

[0041] The processor 21 corrects the fingerprint center position so that the center position Pt11 of the fingerprint becomes the center position Pt12 of the fingerprint image IMG11 (i.e., the field of view) (St15). The processor 21 corrects each of the multiple correction control points Cn01A to Cn100A so that they are arranged at equal intervals at both ends of the X-axis and Y-axis directions, with the center position Pt12 as the center. The processor 21 generates the authentication fingerprint image IMG12 by executing a process to smoothly transform the position of each pixel included in the fingerprint image IMG11 so that the pixels at each coordinate of the correction control points Cn01A to Cn100A are located at the respective coordinates of the correction control points Cn01B to Cn100B after they have been arranged at equal intervals (St16).

[0042] For example, in the example shown in FIG. 4, each of the correction control points Cn01A to Cn100A is a correction control point before correction. Also, each of the correction control points Cn01B to Cn100B is a correction control point after correction. The processor 21 corrects (transforms) the pixels corresponding to the coordinates of the correction control points Cn01A to Cn100A to the positions of the coordinates of the correction control points Cn01B to Cn100B respectively. Note that the symbols of each correction control point are given the same symbol except for the last alphabet in order to make the positional relationship of the correction control points that change before and after correction easier to understand.

[0043] By executing such correction processing, the processor 21 can correct the fingerprint image IMG11 captured in a state where the angle of the fingertip FN11 (fingerprint PTN11) is not substantially parallel to the image sensor of the imaging unit 13 to the authentication fingerprint image IMG12 in which the center position Pt11 of the fingerprint PTN11 is corrected to the center position Pt12 of the fingerprint image IMG11. As shown in FIG. 5, the authentication fingerprint image IMG12 obtained by the correction is an image in which the fingertip FN12 is imaged from the front and the fingerprint PTN12 is imaged at the center position Pt12 of the fingerprint image IMG11.

[0044] The processor 21 executes fingerprint authentication by comparing the generated authentication fingerprint image IMG12 with the registered fingerprint image (St17). The processor 21 outputs the authentication result to the monitor 24 (St18).

[0045] Thereby, the correction device P2 in the first embodiment can correct the orientation of the finger (fingerprint) by cutting out the fingerprint area from the captured image to generate the fingerprint image IMG11. Also, the correction device P2 can correct the orientation and angle of the finger by generating the authentication fingerprint image IMG12 in which the center position Pt11 of the fingerprint is corrected, and can obtain the authentication fingerprint image IMG12 in a state where the finger pad is substantially parallel to the image sensor of the imaging unit 13. As described above, the correction device P2 can more effectively suppress a decrease in the matching accuracy in fingerprint authentication by generating the authentication fingerprint image IMG12 in which the center position of the fingerprint shown in the fingerprint image IMG11 is corrected.

[0046] Also, as described above, the correction device P2 in Embodiment 1 executes correction processing for each fingerprint image. Therefore, the correction device P2 can generate the authentication fingerprint image IMG12 of each finger from a single captured image even when there are variations in the individuality of the user's fingers (e.g., distortion for each finger, etc.) or the orientation and angle of the fingerprint due to the way the finger is held, by capturing the fingerprints of multiple fingers together. Therefore, the correction device P2 can more effectively suppress a decrease in the matching accuracy in fingerprint authentication even when the fingerprints of multiple fingers are captured together.

[0047] (Embodiment 2) The biometric data correction system 100 according to Embodiment 1 shows an example in which the correction device P2 executes correction of a fingerprint image, registration of a registered fingerprint image, and fingerprint authentication. The biometric data correction system 100A according to Embodiment 2 described below will explain an example in which a device (correction device P2A) that executes correction of a fingerprint image and a device (authentication device P2B) that executes registration of a registered fingerprint image and fingerprint authentication are each separately configured.

[0048] Note that the biometric data correction system 100A according to Embodiment 2 has the same configuration and functions as the biometric data correction system 100 according to Embodiment 1. Therefore, in the following description of Embodiment 2, the description of the same configuration and functions as those of the biometric data correction system 100 according to Embodiment 1 will be omitted.

[0049] Referring to FIGS. 6 and 7 respectively, the biometric data correction system 100A according to Embodiment 2 will be described. FIG. 6 is a diagram showing an example of the overall configuration of the biometric data correction system 100A according to Embodiment 2. FIG. 7 is a block diagram showing an example of the internal configuration of the biometric data correction system 100A according to Embodiment 2.

[0050] The biometric data correction system 100A in Embodiment 2 generates a fingerprint image used for fingerprint authentication by correcting the captured image captured by the fingerprint acquisition device P1 by the correction device P2A. The biometric data correction system 100A executes fingerprint authentication using the authentication fingerprint image by the authentication device P2B.

[0051] The biometric data correction system 100A includes, as an example, a fingerprint acquisition device P1, a correction device P2A, an authentication device P2B, a registered fingerprint database DB, and a network NW. Note that the configuration of the biometric data correction system 100A shown in Figures 6 and 7 is an example and is not limited thereto.

[0052] The correction device P2A is connected to the fingerprint acquisition device P1 and the authentication device P2B via wired or wireless communication to perform data transmission and reception. The correction device P2A generates an authentication fingerprint image for use in fingerprint authentication and transmits it to the authentication device P2B. The correction device P2A includes a communication unit 20A, a processor 21A, a memory 22, and a fingerprint acquisition unit 23.

[0053] The communication unit 20A is connected to the fingerprint acquisition device P1 and the authentication device P2B via wireless or wired communication, respectively. The communication unit 20A transmits various data or information received from the processor 21 to the fingerprint acquisition device P1 or the authentication device P2B. The communication unit 20A also outputs various data (e.g., captured images) or information received from the fingerprint acquisition device P1 or the authentication device P2B to the processor 21.

[0054] The processor 21A is configured using, for example, a CPU, FPGA, or GPU, and works in cooperation with the memory 22 to perform various processes and controls. Specifically, the processor 21A refers to the programs and data held in the memory 22 and executes those programs to realize the functions of each part. These parts refer to functions such as the correction unit 211. Furthermore, the method of correcting the fingerprint image IMG 11 by the processor 21A is the same as the operation procedure of the correction device P2 described with reference to Figures 3 to 5, so the explanation is omitted.

[0055] The authentication device P2B is connected to the correction device P2A via wired or wireless communication to transmit and receive data. The authentication device P2B performs biometric authentication using a fingerprint image IMG11 or an authentication fingerprint image IMG12 for use in fingerprint authentication. The authentication device P2B includes a communication unit 30, a processor 31, a memory 32, and a monitor 33. In addition to biometric authentication using the authentication fingerprint image IMG12, the authentication device P2B may also be capable of registering the authentication fingerprint image IMG12 as a registered fingerprint image in a registered fingerprint database DB.

[0056] The communication unit 30 is connected to the correction device P2A via wireless or wired communication. The communication unit 30 transmits various data or information received from the processor 31 to the authentication device P2B. The communication unit 30 also outputs various data (for example, fingerprint image IMG11 or authentication fingerprint image IMG12) or information received from the authentication device P2B to the processor 31.

[0057] The processor 31 is configured using, for example, a CPU, FPGA, or GPU, and works in cooperation with the memory 32 to perform various processes and controls. Specifically, the processor 31 refers to the programs and data held in the memory 32 and executes those programs to realize the functions of each part. These parts include, for example, the registration unit 311 and the authentication unit 312.

[0058] The registration unit 311 is capable of performing the same functions as the registration unit 212 and performs the registration (storage) of fingerprint images of users whose fingerprint images used for biometric authentication have not yet been registered. When the registration unit 311 performs fingerprint image registration, it acquires the authentication fingerprint image IMG12 transmitted from the correction device P2A and registers (stores) it in the registered fingerprint database DB as a registered fingerprint image used for biometric authentication.

[0059] The authentication unit 312 can perform the same functions as the authentication unit 213 and acquires the authentication fingerprint image IMG12 transmitted from the correction device P2A. The authentication unit 312 compares the authentication fingerprint image IMG12 with at least one registered fingerprint image registered in the registered fingerprint database DB and performs fingerprint authentication. The authentication unit 312 outputs the authentication result to the monitor 33. Note that the destination for sending and outputting the authentication result is not limited to the example described above.

[0060] Memory 32 includes, for example, RAM as work memory used when executing each process of the processor 31, and ROM for storing programs and data that define the operation of the processor 31. Data or information generated or acquired by the processor 31 is temporarily stored in RAM. Programs that define the operation of the processor 31 are written in ROM. Memory 32 also stores various data, software, and programs for realizing functions such as the authentication unit 312.

[0061] The monitor 33 is configured using, for example, an LCD or an organic EL display. The monitor 33 outputs and displays various screens that notify the fingerprint image IMG 11, the authentication fingerprint image IMG 12, or the authentication result from the authentication unit 312.

[0062] The network NW performs data transmission and reception between the fingerprint acquisition device P1 and the correction device P2A.

[0063] As described above, the biometric data correction system 100A according to Embodiment 2 enables biometric authentication using a corrected authentication fingerprint image IMG12, even when, for example, the owner of the authentication device P2B that performs fingerprint authentication and the owner of the correction device P2A that corrects the fingerprint image are different people.

[0064] (Note) The following technologies are disclosed based on the descriptions of each embodiment above.

[0065] (Technology 1) A biometric data correction device (correction device P2, P2A) comprising: an acquisition unit (processor 21, 21A) that acquires a fingerprint image IMG11 of a fingerprint captured in a non-contact state; a detection unit (processor 21, 21A) that detects a first center position (center position Pt11) of the fingerprint as captured in the fingerprint image IMG11; a calculation unit (processor 21, 21A) that calculates a plurality of correction control points that correct the first center position (center position Pt11) of the fingerprint to a second center position (center position Pt12) of the fingerprint image IMG11 based on the first center position (center position Pt11) of the fingerprint with respect to the field of view of the fingerprint image IMG11; and a correction unit (processor 21, 21A) that outputs a corrected image (authentication fingerprint image IMG12) in which the coordinates of pixels on the fingerprint image IMG11 have been corrected based on the plurality of correction control points. As a result, the correction devices P2 and P2A generate an authentication fingerprint image IMG12 with the center position Pt11 of the fingerprint corrected, thereby correcting the angle of the finger (fingerprint) and obtaining an authentication fingerprint image IMG12 that appears as if the pad of the finger was captured in a state where it was approximately parallel to the image sensor of the imaging unit 13. Therefore, by generating an authentication fingerprint image IMG12 with the center position of the fingerprint captured in the fingerprint image IMG11 corrected, the correction devices P2 and P2A can more effectively suppress the decrease in matching accuracy in fingerprint authentication.

[0066] (Technology 2) The detection unit (processors 21, 21A) calculates the confidence level of the first central position (central position Pt11) of the fingerprint, and if it is determined that the calculated confidence level is equal to or greater than a threshold, it causes the calculation unit (processors 21, 21A) and the correction unit (processors 21, 21A) to generate the corrected image (authentication fingerprint image IMG12), and if it is determined that the calculated confidence level is not equal to or greater than a threshold, it omits the generation of the corrected image (authentication fingerprint image IMG12) and outputs the fingerprint image IMG11, as described in (Technology 1), a biometric data correction device (correction device P2, P2A). As a result, the correction devices P2, P2A can more effectively suppress the decrease in correction accuracy and the decrease in fingerprint authentication accuracy by determining whether or not to perform the correction process based on the confidence level of the detected central position Pt11 of the fingerprint.

[0067] (Technology 3) The acquisition unit (processors 21, 21A) acquires an image of at least one fingertip, and extracts a region including the first joint of the fingertip from the acquired image to acquire the fingerprint image IMG11 for each finger, as described in (Technology 1) or (Technology 2), a biometric data correction device (correction device P2, P2A). As a result, the correction devices P2, P2A can acquire a fingerprint image of only one fingertip (fingerprint) from an image of multiple fingertips (fingerprints) captured simultaneously. Furthermore, the correction devices P2, P2A can correct the orientation of the fingerprint by extracting the fingerprint region from the image to generate the fingerprint image IMG11.

[0068] (Technology 4) The detection unit (processors 21, 21A) estimates the pattern of the fingerprint and detects the first central position (central position Pt11) of the fingerprint based on the estimated pattern, as described in any one of (Technology 1) to (Technology 3), a biometric data correction device (correction device P2, P2A). As a result, the correction devices P2, P2A can improve the detection accuracy of the central position Pt11 of the fingerprint. Therefore, the correction devices P2, P2A can correct the correction accuracy of the fingerprint image IMG11, that is, the angle of the finger (fingerprint), with higher precision, and thus can generate an authentication fingerprint image IMG12 that is more suitable for fingerprint authentication.

[0069] (Technical 5) A biometric data correction method performed by at least one processor (processors 21, 21A), comprising: acquiring a fingerprint image IMG11 of a fingerprint captured in a non-contact state; detecting a first center position (center position Pt11) of the fingerprint as depicted in the fingerprint image IMG11; calculating a plurality of correction control points that correct the first center position (center position Pt11) of the fingerprint to a second center position (center position Pt12) of the fingerprint image IMG11 based on the first center position (center position Pt11) of the fingerprint relative to the field of view of the fingerprint image IMG11; and outputting a corrected image (authentication fingerprint image IMG12) in which the coordinates of pixels on the fingerprint image IMG11 have been corrected based on the plurality of correction control points. As a result, the correction devices P2 and P2A generate an authentication fingerprint image IMG12 with the center position Pt11 of the fingerprint corrected, thereby correcting the angle of the finger (fingerprint) and obtaining an authentication fingerprint image IMG12 that appears as if the pad of the finger was captured in a state where it was approximately parallel to the image sensor of the imaging unit 13. Therefore, by generating an authentication fingerprint image IMG12 with the center position of the fingerprint captured in the fingerprint image IMG11 corrected, the correction devices P2 and P2A can more effectively suppress the decrease in matching accuracy in fingerprint authentication.

[0070] (Technical 6) A biometric data correction system 100, 100A comprising: an imaging device (fingerprint acquisition device P1) for capturing the fingerprint of a person to be authenticated (user) in a non-contact state; and correction devices P2, P2A that can communicate with the imaging device (fingerprint acquisition device P1), wherein the imaging device (fingerprint acquisition device P1) transmits an image of at least one fingertip to the correction devices P2, P2A; the correction devices P2, P2A extract a region including the first joint of the fingertip (fingerprint region) from the image to generate a fingerprint image IMG 11 of the fingerprint; and detect the first center position (center position Pt 11) of the fingerprint as depicted in the fingerprint image IMG 11. The biometric data correction system 100, 100A calculates a plurality of correction control points that correct the first center position (center position Pt11) of the fingerprint to the second center position (center position Pt12) of the fingerprint image IMG11 based on the field of view of the fingerprint image IMG11, and outputs a corrected image (authentication fingerprint image IMG12) in which the coordinates of the pixels on the fingerprint image IMG11 have been corrected based on the plurality of correction control points. As a result, the biometric data correction system 100, 100A can correct the angle of the finger (fingerprint) by generating an authentication fingerprint image IMG12 in which the center position Pt11 of the fingerprint has been corrected, and can obtain an authentication fingerprint image IMG12 that appears as if the pad of the finger was captured in a state where it was approximately parallel to the image sensor of the imaging unit 13. Therefore, the biometric data correction systems 100 and 100A can more effectively suppress the decrease in matching accuracy in fingerprint authentication by generating an authentication fingerprint image IMG 12 in which the central position of the fingerprint captured in the fingerprint image IMG 11 has been corrected.

[0071] Although various embodiments have been described above with reference to the attached drawings, this disclosure is not limited to such examples. It will be clear to those skilled in the art that various modifications, alterations, substitutions, additions, deletions, and equivalents can be conceived within the scope of the claims, and these will also be understood to fall within the technical scope of this disclosure. Furthermore, the components of the various embodiments described above can be combined arbitrarily without departing from the spirit of the invention.

[0072] This application is based on a Japanese patent application (Patent Application No. 2024-193560) filed on November 5, 2024, the contents of which are incorporated herein by reference.

[0073] This disclosure is useful as a presentation of a biometric data correction device, a biometric data correction method, and a biometric data correction system for correcting the angles of biological parts included in biological images acquired in free space.

[0074] 10, 20, 20A, 30 Communication unit 11, 21, 21A, 31 Processor 12, 22, 32 Memory 13 Imaging unit 23 Fingerprint acquisition unit 24, 33 Monitor 100, 100A Biometric data correction system 211 Correction unit 212, 311 Registration unit 213, 312 Authentication unit Cn01A, Cn01B Correction control point Cn100A, Cn100B Correction control point DB Registered fingerprint database FN11, FN12 Fingertip IMG11 Fingerprint image IMG12 Authentication fingerprint image NW Network P1 Fingerprint acquisition device P2, P2A Correction device P2B Authentication device Pt11, Pt12 Center position PTN11, PTN12 Fingerprint

Claims

1. A biometric data correction device comprising: an acquisition unit that acquires a fingerprint image of a fingerprint captured in a non-contact state; a detection unit that detects a first central position of the fingerprint as depicted in the fingerprint image; a calculation unit that calculates a plurality of correction control points that correct the first central position of the fingerprint to a second central position of the fingerprint image based on the first central position of the fingerprint with respect to the field of view of the fingerprint image; and a correction unit that outputs a corrected image in which the coordinates of pixels on the fingerprint image have been corrected based on the plurality of correction control points.

2. The biometric data correction device according to claim 1, wherein the detection unit calculates the confidence level of the first central position of the fingerprint, and if it determines that the calculated confidence level is equal to or greater than a threshold, it causes the calculation unit and the correction unit to generate the corrected image, and if it determines that the calculated confidence level is not equal to or greater than a threshold, it omits the generation of the corrected image and outputs the fingerprint image.

3. The biometric data correction device according to claim 1, wherein the acquisition unit acquires an image of at least one fingertip, and extracts a region including the first joint of the fingertip from the acquired image to acquire a fingerprint image for each finger.

4. The biometric data correction device according to claim 1, wherein the detection unit estimates the pattern of the fingerprint and detects the first central position of the fingerprint based on the estimated pattern.

5. A biometric data correction method performed by at least one processor, comprising: acquiring a fingerprint image of a fingerprint captured in a non-contact state; detecting a first central position of the fingerprint as depicted in the fingerprint image; calculating a plurality of correction control points that correct the first central position of the fingerprint to a second central position of the fingerprint image based on the first central position of the fingerprint with respect to the field of view of the fingerprint image; and outputting a corrected image in which the coordinates of pixels on the fingerprint image have been corrected based on the plurality of correction control points.

6. A biometric data correction system comprising: an imaging device for capturing a fingerprint of a person to be authenticated in a non-contact state; and a correction device capable of communicating with the imaging device, wherein the imaging device transmits an image of at least one fingertip to the correction device; the correction device extracts a region including the first joint of the fingertip from the image to generate a fingerprint image of the fingerprint; detects a first central position of the fingerprint as depicted in the fingerprint image; calculates a plurality of correction control points that correct the first central position of the fingerprint to a second central position of the fingerprint image based on the first central position of the fingerprint with respect to the field of view of the fingerprint image; and outputs a corrected image in which the coordinates of pixels on the fingerprint image have been corrected based on the plurality of correction control points.