WO2018207571A1 - 認証情報処理プログラム及び認証情報処理装置 - Google Patents
認証情報処理プログラム及び認証情報処理装置 Download PDFInfo
- Publication number
- WO2018207571A1 WO2018207571A1 PCT/JP2018/015999 JP2018015999W WO2018207571A1 WO 2018207571 A1 WO2018207571 A1 WO 2018207571A1 JP 2018015999 W JP2018015999 W JP 2018015999W WO 2018207571 A1 WO2018207571 A1 WO 2018207571A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- reference point
- peripheral
- authentication information
- reference points
- Prior art date
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/117—Identification of persons
- A61B5/1171—Identification of persons based on the shapes or appearances of their bodies or parts thereof
- A61B5/1172—Identification of persons based on the shapes or appearances of their bodies or parts thereof using fingerprinting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/45—Structures or tools for the administration of authentication
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/70—Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer
- G06F21/82—Protecting input, output or interconnection devices
- G06F21/83—Protecting input, output or interconnection devices input devices, e.g. keyboards, mice or controllers thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
- G06T7/0014—Biomedical image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1347—Preprocessing; Feature extraction
- G06V40/1353—Extracting features related to minutiae or pores
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1365—Matching; Classification
- G06V40/1371—Matching features related to minutiae or pores
Definitions
- the present invention relates to an authentication information processing program and an authentication information processing apparatus for analyzing information and creating information used for collation of skin information.
- the biometric identification device of Patent Document 1 uses a fingerprint ridge and valley pattern extracted from biometric information and a sweating hole extracted from biometric information to verify authentication for registration information. Authenticate information.
- the biometric identification device described in Patent Document 1 has not been sufficiently studied from the viewpoint of improving the authentication speed.
- An object of the present invention is to provide an authentication information processing program and an authentication information processing apparatus that are capable of generating information used for skin authentication, which contributes to an improvement in authentication speed as compared with the prior art.
- An authentication information processing program includes an image acquisition step of acquiring an image on a computer including a processor and a storage device, and a ridge on the skin from the image acquired in the image acquisition step.
- a reference point representing a perspiration hole, a reference point determining step for acquiring position information that is information corresponding to the position of the reference point on the image, and a plurality of the reference points acquired in the reference point determining step When one of the points is selected as a center reference point, the distance from the center reference point is less than a predetermined value, and between the adjacent ridges of the skin that are between the center reference point
- a peripheral reference point extracting step of extracting a predetermined number of reference points having a number of valleys of 1 or less as peripheral reference points based on a predetermined condition; and for each of the plurality of reference points, the position information of the central reference point
- an information generation step of generating sweat pore related information in association with attribute information representing a feature of the arrangement of each of the predetermined number of the
- the authentication information processing program can generate sweat hole related information based on an image on a computer and store it in a storage device.
- the placement of sweat pores on the skin ridge is as unique as fingerprints and voiceprints and is said to remain unchanged throughout life. There is a possibility that a plurality of sweat holes can be acquired even when the size of the image representing the skin is smaller than the conventional size and the branch points and end points of the ridges are not included in the image.
- the reference point represents a sweat hole on the ridge
- the attribute information represents a feature of the arrangement of each of the predetermined peripheral reference points on the image.
- the attribute information is information that represents the feature of the arrangement of a predetermined number of sweat holes around the target sweat hole, and can be said to be information that emphasizes the feature portion of the biological information represented by the image.
- the attribute information includes, for example, verification authentication information used for calculating the similarity of authentication information, reference points for verification used for processing to determine correspondence between the authentication information for registration, and reference points for registration. Can be suitably used for the process of extracting. Therefore, by executing the authentication information processing program of the first aspect, the computer can generate sweat hole related information that is information used for skin authentication and contributes to an improvement in authentication speed as compared with the conventional case.
- the reference point determination step may determine the center of gravity of the sweat hole in the image as the reference point and acquire the position information of the reference point.
- the computer can determine the center of gravity of the sweat hole representing the feature of the shape and size of the sweat hole as a reference point and generate sweat hole related information.
- the predetermined condition includes a line segment connecting the peripheral reference point that has already been selected and the central reference point, the reference point that is a candidate for the peripheral reference point, and the center.
- the condition may be that the predetermined number of the reference points whose angle formed by the line segment connecting the reference points is equal to or greater than a predetermined angle is selected in the order of the distance from the central reference point.
- the attribute information includes information indicating the number of the peripheral reference points on the ridge that are the same as the central reference point among the predetermined peripheral reference points, and the central reference point And each of the predetermined number of line segments when connecting each of the predetermined number of peripheral reference points as a radiation component, and the peripheral reference points on the two adjacent radiation components are sequentially arranged around the central reference point.
- Classification information including information indicating the number of the peripheral lines on the ridge when each of the predetermined number of connected line segments is a peripheral line may be included.
- the computer uses the verification information used for the process of determining the correspondence between the verification authentication information used for calculating the similarity of the authentication information and the registration authentication information. Classification information that can contribute to speeding up the process of extracting the reference point and the registration reference point can be generated.
- the attribute information may include radiation information that is information indicating whether each of the peripheral reference points indicates whether the central reference point and the peripheral reference point are on the same ridge. Good.
- the computer uses the verification information used for the process of determining the correspondence between the verification authentication information used for calculating the similarity of the authentication information and the registration authentication information. Radiation information that can contribute to speeding up the process of extracting the reference point and the registration reference point can be generated.
- the attribute information is adjacent to each other with a predetermined number of line segments when the center reference point and each of the predetermined number of peripheral reference points are connected as radiation. Whether each of the predetermined line segments connecting the peripheral reference points on the two radiation components in turn around the central reference point is a peripheral line segment, whether the peripheral line segments are on the same ridge Circulation information that is information shown for each of the peripheral reference points that are the starting points for the circuit line may be included.
- the computer uses the verification information used for the process of determining the correspondence between the verification authentication information used for calculating the similarity of the authentication information and the registration authentication information. Circulation information that can contribute to speeding up the process of extracting the reference point and the reference point for registration can be generated.
- the information generation step includes, for each of the plurality of reference points, in addition to the position information of the central reference point and the attribute information, the predetermined number of peripheral reference points.
- the sweat pore related information including peripheral information that is information based on the position information may be generated.
- the computer can use the peripherals that can be used for the process of determining the correspondence between the authentication information for verification used for calculating the similarity of skin authentication and the authentication information for registration. Sweat related information including information can be generated.
- the information generation step generates, as the peripheral information, the sweat pore related information including a distance from the central reference point for each of the predetermined peripheral reference points. May be.
- the computer can use a distance that can be used for processing for determining the correspondence between the authentication information for verification used for calculating the similarity of skin authentication and the authentication information for registration. Can be generated.
- the computer can reduce the time required for skin authentication as compared with the case where the distance is calculated every time skin authentication is performed.
- the information generation step includes a predetermined direction of a line segment connecting the central reference point and the peripheral reference point for each of the predetermined peripheral reference points as the peripheral information.
- the sweat hole related information including an angle with respect to may be generated.
- the computer can use the angle that can be used for the process of determining the correspondence between the authentication information for verification used for calculating the similarity of the authentication information and the authentication information for registration. Can be generated.
- the computer can reduce the time required for skin authentication as compared with the case of calculating the angle every time skin authentication is performed.
- the information generation step matches the arrangement order of the peripheral information with the arrangement order of the attribute information, and is the same as the central reference point among the predetermined peripheral reference points
- the separation reference point that is on the ridge and has the longest distance from the center reference point as the arrangement order of the separation reference points having the longest distance, and is centered on the center reference point based on the arrangement on the image
- the second and subsequent arrangement orders may be set around a predetermined direction.
- the computer displays the sweat hole related information in which the arrangement order is set based on the combination of the center reference point and the peripheral reference point that are considered to be the least affected by the distortion of the image. Can be generated.
- the verification authentication information generated in the information generation step and the registration authentication information stored in the storage device are associated with the reference point. Further, a pair of the reference point for matching and the reference point for registration that match the attribute information is a candidate for comparison of the correspondence between the authentication information for matching and the authentication information for registration.
- the computer determines the correspondence between the authentication information for verification used for calculating the similarity and the authentication information for registration by using all the reference points included in the authentication information. It can be determined in a shorter time compared to the case where the determination is made by comparison.
- the verification authentication information generated in the information generation step and the registration authentication information stored in the storage device are associated with the reference point. Further, a pair of the reference point for matching and the reference point for registration that match the attribute information is a candidate for comparison of the correspondence between the authentication information for matching and the authentication information for registration.
- a pair candidate extraction step for extracting as a pair candidate, and for the pair candidate extracted in the pair candidate extraction step, the peripheral information in the sweat pore related information is compared with the authentication information for verification and the registration
- An instruction for executing a correspondence determining step for determining a correspondence with the authentication information may be further included.
- the computer determines the correspondence between the authentication information for verification used for calculating the similarity and the authentication information for registration by using all the reference points included in the authentication information. It can be determined in a shorter time compared to the case where the determination is made by comparison.
- the computer can determine the correspondence between the verification authentication information and the registration authentication information by comparing the peripheral information of the pair candidates.
- the authentication information for verification using the correspondence between the authentication information for verification determined in the correspondence determination step and the authentication information for registration, and the registration information You may further include the instruction
- An authentication information processing apparatus includes a processor and a storage device, and the processor acquires an image, an image acquisition unit that acquires an image, and a sweat hole on a skin ridge from the image acquired by the image acquisition unit. And a reference point determination unit that acquires position information that is information corresponding to the position of the reference point on the image, and a plurality of the reference points acquired by the reference point determination unit.
- the distance from the center reference point is less than a predetermined value, and the valley between the adjacent ridges of the skin between the center reference point and Extracting means for extracting a predetermined number of reference points having a number of 1 or less as peripheral reference points based on a predetermined condition; for each of the plurality of reference points, the position information of the central reference point; and the extracting means Extracted said Information generating means for generating sweat hole-related information in association with attribute information representing the feature of the arrangement of each of the fixed peripheral reference points on the image, and the information generating means It functions as storage control means for storing the perforation-related information relating to each in the storage device as authentication information used for skin authentication.
- the technology disclosed in the present specification can be realized in various forms, for example, a non-transitory computer-readable medium storing the authentication information processing program of the first aspect, and the authentication of the second aspect. It can be realized in the form of a method for realizing the functions of the information processing apparatus.
- FIG. 2 is a block diagram of the device 10.
- FIG. 3 is a functional block diagram of the device 10.
- FIG. It is a flowchart of authentication information processing of a first embodiment. It is explanatory drawing of the process which extracts a reference point based on the image. It is explanatory drawing of the process in which the periphery reference point a0 to a7 is determined about the reference point A1 based on the image 45, and sweat hole related information is acquired. (A) to (J) are explanatory diagrams of pattern examples of attribute information. It is explanatory drawing of the authentication information 33 containing the sweat hole related information 30.
- FIG. It is a flowchart of the authentication information processing of 2nd embodiment. It is a flowchart of the image analysis process performed by the authentication information processing of FIG.
- FIG. 10 is a flowchart of sweat hole related information generation processing executed in the image analysis processing of FIG. 9. It is explanatory drawing of the process which acquires sweat hole related information about center reference point A1. It is explanatory drawing of the sweat hole relevant information 50 acquired about center reference point A1. It is explanatory drawing of the sweat hole related information acquired about each of several reference points including center reference point A1 to A10 based on the image 45. FIG. It is explanatory drawing of the authentication information 70 containing the sweat hole relevant information acquired about each of several reference points including center reference point A1 to A10 based on the image 45. FIG. It is explanatory drawing of the sweat pore related information acquired about each of several reference points including center reference point B1 to B9 based on the image 65. FIG.
- 18 is a table showing a combination of a reference point for matching extracted as a pair candidate and a reference point for registration in the pair candidate extraction process of FIG. 17, and the score and rotation angle of the pair candidate. It is a histogram which classified the rotation angle of a pair candidate into the angle of the predetermined range, and is explanatory drawing of the process which narrows down a pair candidate by a rotation angle. It is explanatory drawing of the process which narrows down a pair candidate with the positional relationship of two sets of pair candidates. 6 is a graph showing the results of evaluation test 1. 10 is a graph showing the results of evaluation test 3.
- An authentication information processing apparatus (hereinafter simply referred to as “apparatus”) 10 is an electronic device having a function of generating authentication information used for verification from biometric information (skin information).
- the skin information is selected from biological information represented by an image of hairless skin such as a finger, palm, and sole. Skin information in this example is fingerprints and sweat holes.
- the device 10 of this example is a known smartphone.
- the apparatus 10 analyzes an image obtained by photographing a fingerprint and a sweat hole, generates authentication information for registration necessary for collation using skin information, and stores a database (DB) 28 stored in the flash memory 4 of the apparatus 10.
- the function to memorize is provided.
- the apparatus 10 analyzes an image obtained by photographing a fingerprint and a sweat hole, generates authentication information for verification necessary for verification using skin information, and stores the generated verification authentication information and the DB 28.
- a function for determining the correspondence with the authentication information for registration is provided.
- the device 10 includes a CPU 1, a ROM 2, a RAM 3, a flash memory 4, a communication I / F 5, a display unit 6, a touch panel 7, and a biological information acquisition device 8.
- the CPU 1 is a processor that controls the apparatus 10.
- the CPU 1 is electrically connected to the ROM 2, RAM 3, flash memory 4, communication I / F 5, display unit 6, touch panel 7, and biological information acquisition device 8.
- the ROM 2 stores a BIOS, a boot program, and initial setting values.
- the RAM 3 stores various temporary data.
- the flash memory 4 stores a program executed by the CPU 1 to control the device 10, an OS (Operating System), and a DB 28.
- the communication I / F 5 is a controller for executing communication with an external device.
- the display unit 6 is a liquid crystal display.
- the touch panel 7 is provided on the surface of the display unit 6.
- the biological information acquisition device 8 acquires an image obtained by photographing the skin.
- the biometric information acquisition apparatus 8 of this example is an optical area type sensor or a microscope, and expresses color information with a gradation value of 256 gradations for each pixel.
- the color information is information representing a color.
- the resolution of the image is preferably 800 dpi (dots per inch) or more.
- the resolution of the biometric information acquisition apparatus 8 of this example is 2000 dpi as an example.
- the apparatus 10 includes a biological information acquisition device 8, an image acquisition unit 21, a reference point determination unit 22, an extraction unit 23, an information generation unit 24, a registration unit 26, a collation unit 27, and a DB 28.
- the processing corresponding to each functional block is executed by the CPU 1 (see FIG. 1).
- the biological information acquisition apparatus 8 outputs an image to the image acquisition unit 21.
- the image acquisition unit 21 acquires the image output from the biological information acquisition device 8 (S1).
- the biological information acquisition device 8 acquires the image 41.
- the image 41 is a diagram schematically showing a part of a rectangular image having, for example, a resolution of 2000 dpi, an X direction (left-right direction) of 480 pixels, and a Y direction (up-down direction) of 800 pixels.
- the reference point determination unit 22 determines a reference point representing a sweat hole on the ridge of the skin from the image 41 acquired in S1, and acquires position information that is information corresponding to the position of the reference point on the image (S2). ).
- the reference point determination unit 22 determines the center of gravity of the sweat hole in the image as a reference point, and acquires position information of the reference point.
- the reference point position information is represented, for example, by two-dimensional coordinates in the image coordinate system.
- the two-dimensional coordinates of the image coordinate system of this example are coordinates set in units of pixels based on the positions of the pixels in the image.
- the CPU 1 sets the position of the upper left pixel of the image 41 as the origin of the two-dimensional coordinate 46 in the image coordinate system.
- the position of a pixel that is separated by x pixels in the X plus direction from the origin of the two-dimensional coordinate 46 and separated by y pixels in the Y plus direction from the origin is denoted as coordinates (x, y).
- the reference point determination unit 22 determines the reference point by the following procedure, for example. As illustrated in FIG. 4, the reference point determination unit 22 generates an image 42 that represents a ridge and an image 43 that represents a figure of a closed region including sweat pores from the image 41 acquired in S ⁇ b> 1.
- the image 42 is obtained by binarizing the image 41, for example.
- the image 42 is obtained by providing the image 41 to a plurality of image processing filters used for processing of the minutiae method.
- the black portion represents a ridge
- the white portion represents a valley portion between the ridges.
- the image 43 is obtained, for example, by applying the image 41 to an image processing filter that can extract a gray value portion within a predetermined range.
- Sweat holes are located on the ridges.
- the reference point determination unit 22 compares the image 42 and the image 43 with each other as shown in the image 44 and compares the images 42 and 43 with a closed region portion such as a circle or a hook disposed on the ridge indicated by the black portion. As specified.
- the reference point determination unit 22 determines the area center of gravity of the identified sweat hole as a reference point representing the sweat hole.
- the reference point determination unit 22 scans the image 45 in order from left to right and from top to bottom, and assigns an ID to each reference point. In the process of S2, for example, a plurality of reference points including the reference points P1 to P12 of the image 45 are extracted. For each reference point, a list 47 in which the assigned ID is associated with the two-dimensional coordinates of the image coordinate system is created.
- the extraction unit 23 selects one of the plurality of reference points acquired in S2 as the center reference point, the distance from the center reference point is less than a predetermined value and is between the center reference point
- a predetermined number of reference points whose number of valleys between adjacent ridges of the skin is 1 or less are extracted as peripheral reference points based on a predetermined condition (S3).
- the number of valleys between the adjacent ridges of the skin between the center reference point and the center reference point is specified by, for example, the number of continuous white areas in the image 42 that exist between the center reference point.
- the predetermined distance, the predetermined number, and the predetermined condition are determined in consideration of the distance between the sweat holes, the image resolution, the authentication accuracy, and the like.
- the predetermined distance is preferably larger than the average value H of the distance between sweat holes and smaller than three times the average value H.
- the predetermined number is preferably 4 or more and 15 or less.
- the predetermined distance is the radius R
- the predetermined number is eight
- the predetermined condition is that the line segment connecting the already selected peripheral reference point and the center reference point
- the reference points that are candidates for the peripheral reference point 5 is a condition for selecting a predetermined number of reference points in the order of the distance from the central reference point, the angle of the angle formed by the line connecting the center reference point and the central reference point is equal to or greater than the predetermined angle.
- the points a0 to a7 shown are extracted as the peripheral reference points.
- the predetermined angle is determined in consideration of the distance between sweat holes, the resolution of the image, the number of peripheral reference points, authentication accuracy, and the like.
- the predetermined number is 8
- the predetermined angle is preferably selected in the range of 5 degrees to 45 degrees, and is 15 degrees in this example.
- the process of S3 is executed using each of the plurality of reference points acquired in the process of S2 as a central reference point in order.
- the information generation unit 24 is an attribute representing the positional information of the center reference point and the characteristics of the arrangement of each of the predetermined number of peripheral reference points extracted in S3 on the image.
- the sweat hole related information in association with the information is generated (S4).
- the attribute information may be information that represents the feature of the arrangement of the peripheral reference points on the image with respect to the central reference point.
- the attribute information is, for example, information set from the viewpoint of whether or not an attention reference point of interest among peripheral reference points is on the same ridge as the central reference point or a predetermined peripheral reference point.
- the attribute information may be information set from the viewpoint of at least one of the distance and angle between the reference point of interest and the center reference point or a predetermined peripheral reference point.
- the attribute information may be information representing one type of feature, or may be a plurality of information representing each of a plurality of types of features.
- Attribute information may include classification information, for example.
- the classification information includes first information indicating the number of peripheral reference points on the same ridge as the central reference point among the predetermined number of peripheral reference points.
- the classification information is that each of the predetermined number of line segments when the central reference point and each of the predetermined peripheral reference points are connected is a radiation component, and the peripheral reference point on two adjacent radiation components is the central reference point.
- Second information indicating the number of peripheral lines on the ridge is included when each of a predetermined number of line segments connected in order around is a peripheral line.
- the radiation components on the same ridge as the center reference point A1 are indicated by solid lines.
- the amount of radiation straddling the valley between the two is indicated by a dashed line.
- the number of peripheral reference points on the same ridge as the center reference point is two.
- the center reference point A1 is on the ridge among the eight peripheral lines connecting the eight peripheral reference points a0 to a7 around the central reference point in order of the peripheral reference points on the two adjacent radiation components.
- the peripheral line is indicated by a first dotted line, and the peripheral line across the valley is indicated by a second dotted line that is finer than the first dotted line. Accordingly, the number of line segments on the ridge among the eight peripheral lines is four. Accordingly, the classification information is represented as “0x24”, for example.
- the attribute information includes classification information, as shown in FIGS. 6A to 6J, the arrangement of the peripheral reference points with respect to the central reference point, as exemplified by the arrangement of the peripheral reference points with respect to the central reference point and the classification information. Different attribute information (classification information) is obtained according to the feature. As illustrated in FIG. 7, the information generation unit 24 generates sweat hole related information 30 by associating position information 31 and attribute information 32 for each reference point.
- the registration unit 26 stores the sweat pore related information 30 of each of the plurality of reference points generated in S4 in the DB 28 as registration authentication information 33 used for collation (S5).
- the collation unit 27 stores the sweat pore-related information generated in S4 in the RAM 3 as collation authentication information used for collation.
- the collation unit 27 collates the verification authentication information with the registration authentication information stored in the DB 28 and performs skin authentication.
- the device 10 ends the process.
- the CPU 1 (image acquisition unit 21) that executes S1 is an example of the image acquisition means of the present invention.
- the process of S2 is an example of the reference point determination step of the present invention.
- the CPU 1 (reference point determination unit 22) that executes S2 is an example of the reference point determination means of the present invention.
- the process of S3 is an example of the peripheral reference point extraction step of the present invention.
- the CPU 1 (extraction unit 23) that executes S3 is an example of the extraction means of the present invention.
- the process of S4 is an example of the information generation step of the present invention.
- the CPU 1 (information generation unit 24) that executes S4 is an example of the information generation means of the present invention.
- the process of S5 is an example of the storage control step of the present invention.
- CPU1 (registration part 26, collation part 27) which performs S5 is an example of the memory
- the authentication information processing is started when the user inputs a start instruction.
- the start instruction includes an instruction on whether the authentication information acquired from the image is registered in the DB 28 as registration authentication information or whether the acquired authentication information is checked against the registration authentication information registered in the DB 28. Including.
- the CPU 1 of the apparatus 10 detects the input of the authentication information processing start instruction, the CPU 1 reads out the authentication information processing program for executing the authentication information processing stored in the flash memory 4 to the RAM 3 and includes the instruction included in the authentication information processing program.
- each step described below is executed according to the above.
- feedback processing that prompts re-input is executed until skin information that satisfies the requirement for extracting the reference point (for example, the sharpness of the image) is acquired.
- An image acquired by authentication information processing satisfies the requirement to extract authentication information from an image using an algorithm.
- Information and data acquired or generated in the course of processing are stored in the RAM 3 as appropriate.
- Various setting values necessary for processing are stored in the flash memory 4 in advance.
- the step is abbreviated as “S”.
- the CPU 1 executes image analysis processing (S11).
- the image analysis process will be described with reference to FIG.
- the biometric information acquisition device 8 detects a finger contact
- the biometric information acquisition device 8 outputs to the CPU 1 a signal that can identify an image obtained by photographing a fingerprint and a sweat hole.
- the CPU 1 receives a signal output from the biological information acquisition device 8.
- the CPU 1 acquires an image based on the received signal (S21).
- S21 for example, an image 41 shown in FIG. 4 is acquired.
- two-dimensional coordinates 46 of an image coordinate system indicated by X and Y are set.
- the CPU 1 binarizes the image 41 acquired in S21, and acquires an image 42 representing a ridge (S22).
- the CPU1 determines a reference point (S23).
- the CPU 1 provides the image 41 to an image processing filter that can extract a gray value portion of a predetermined range, obtains an image 43, and compares the image 42 and the image 43 as shown in an image 44, and compares them.
- a closed region portion such as a circular shape or a hook shape arranged on a ridge indicated by a black portion is specified as a sweat hole.
- the CPU 1 determines the area center of gravity of the identified sweat hole as a reference point representing the sweat hole.
- the CPU 1 assigns an ID to a reference point determined by scanning the image 45 in order from left to right and from top to bottom, and acquires position information.
- the position information in this example is the coordinates in pixel units of the two-dimensional coordinates 46 in the image coordinate system.
- the CPU 1 creates a list 47 representing the correspondence between the assigned ID and the position information, and stores it in the RAM 3.
- the CPU 1 may determine whether or not to identify as a sweat hole by appropriately considering the size and shape of the closed region portion.
- a plurality of reference points including reference points P1 to P12 indicated by black circles on the image 45 are determined.
- the CPU 1 determines whether or not the number of reference points determined in S23 is greater than 0 (S24). When the number of reference points determined in S23 is 0 (S24: NO), the CPU 1 ends the image analysis process and returns the process to the authentication information process of FIG.
- the CPU 1 executes sweat hole related information generation processing (S25). In the sweat hole related information generation process, for each of the plurality of reference points, a process of generating sweat hole related information in which the position information of the center reference point and the attribute information are associated is executed.
- the CPU 1 initializes a list for storing the sweat hole related information (S31).
- the CPU 1 selects one reference point as the center reference point PA from among the plurality of reference points determined in the process of S23 of FIG. 8 (S32).
- the CPU 1 may not select the reference point in the range where it is determined that the peripheral reference point cannot be extracted as the central reference point PA in consideration of the acquisition conditions of the peripheral reference point.
- the CPU 1 may not select a reference point in a range in which the distance from the end of the image 45 is less than half of the predetermined value R described later as the center reference point.
- the reference point A1 of the image 45 shown in FIG. 5 is selected as the center reference point.
- the CPU 1 refers to the list 47 and selects a range in which candidates for the peripheral reference point PB are extracted from the plurality of reference points (S33). Specifically, the CPU 1 sets the value of the Y coordinate of the two-dimensional coordinates among the reference points stored in the list 47 within the range of plus or minus R of the Y coordinate of the center reference point PA selected in the process of S32. This range is a range for extracting candidates for the peripheral reference point PB. In this example, the CPU 1 selects a reference point whose distance from the center reference point PA (for example, Euclidean distance) is within a predetermined value R from among a plurality of reference points determined in the process of S23, for example. In order to obtain PK, first, a range for extracting candidates is selected using the Y coordinate of the list 47.
- the CPU 1 selects one reference point (target reference point) that has not been selected in S34 among the reference points in the range selected in the process of S33, and acquires position information (S34).
- the position information is represented by two-dimensional coordinates 46 in the image coordinate system.
- the CPU 1 calculates the distance between the center reference point and the reference point of interest based on the position information acquired in S34 and the position information of the center reference point selected in S32 (S35).
- the distance calculation method may be set as appropriate.
- the CPU 1 in this example calculates the Euclidean distance between the center reference point and the attention reference point. For example, distances D0 to D7 are calculated for reference points a0 to a7 in FIG.
- the CPU 1 calculates an angle representing the arrangement of the reference point of interest whose position information has been acquired in S34 with respect to the center reference point selected in S32 (S36). An angle calculation method may be set as appropriate.
- the CPU 1 of this example calculates an angle with respect to the reference of the line segment connecting the center reference point and the reference reference point.
- the reference J is the X plus direction (rightward) from the center reference point in the two-dimensional coordinates of the image coordinate system.
- the CPU 1 sets the X plus direction as 0 degrees, sets the clockwise angle from the reference starting from the center reference point as a positive angle, and sets the counterclockwise angle from the reference as a negative angle.
- the CPU 1 represents the angle of the attention reference point as an angle from ⁇ 180 degrees to 180 degrees according to the position of the attention reference point with respect to the center reference point. For example, a positive angle An4 is calculated for the reference point a4, and a negative angle An3 is calculated for the reference point a3.
- the angle may be expressed by other methods such as 0 to 360 degrees.
- the CPU 1 determines whether or not position information has been acquired in the process of S34 for all the reference points in the range selected in S33 (S37). When the process of S34 is not executed for some of the reference points in the range selected in S33 (S37: NO), the CPU 1 returns the process to S34.
- the CPU 1 selects a reference reference point whose distance calculated in S35 is equal to or less than the predetermined value R. Let it be a candidate reference point PK. For example, the CPU 1 sets each of the 14 reference points a0 to a13 located inside the circle C with the radius R centering on the center reference point A1 shown in FIG. 5 as the candidate reference points PK.
- the CPU 1 determines whether or not the number of candidate reference points PK is smaller than the predetermined number 8 (S38). When the number of candidate reference points PK is smaller than 8 (S38: YES), the CPU 1 advances the process to S50 described later. In this case, sweat hole related information for the center reference point PA selected in S32 is not generated. If the number of candidate reference points PK is not smaller than 8 (S38: NO), the CPU 1 has the largest distance calculated in S35 among the reference points a0 to a13 which are candidate reference points PK selected in S38. A reference point a2 that is a small reference point is acquired as a peripheral reference point.
- the CPU 1 acquires, from the candidate reference points PK, a reference point that is the Nth closest to the center reference point PA calculated in the process of S35 (S39). Since the reference point closest to the center reference point PA has already been acquired as the peripheral reference point, the initial value of N is 2. When N is 2, the CPU 1 acquires a point a4 having the second closest distance from the reference points a0 to a13.
- the CPU 1 sets a line segment connecting the peripheral reference point and the central reference point based on the angle calculated in S36 for each reference point already acquired as the peripheral reference point among the candidate reference points PK, and in S39. It is determined whether the absolute value of the angle formed by the line segment formed by the acquired reference point and the center reference point is 15 degrees or more (S40). For example, when there are a plurality of peripheral reference points that have already been determined, the CPU 1 newly acquires the radiation component that connects the central reference point and the peripheral reference point for all of the plurality of peripheral reference points, and is newly acquired in the process of S39.
- a case where the absolute value of the angle between the reference point and the radiation amount of the central reference point is 15 degrees or more is determined as a case where the absolute value of the angle is 15 degrees or more. That is, the CPU 1 determines the radiation component connecting the central reference point and the peripheral reference point, and the reference point and the central reference point newly acquired in the process of S39, even if one of the peripheral reference points has already been determined. If the absolute value of the angle with respect to the radiation component is less than 15 degrees, it is not determined in S40 that the absolute value of the angle is 15 degrees or more.
- the CPU 1 obtains a line segment connecting the reference points A1 and a2, and the reference points A1 and a4, as shown in FIG.
- the CPU 1 adds the reference point acquired in the process of S39 to the peripheral reference points of the central reference point selected in S32 (S41), and there are eight reference points already acquired as the peripheral reference points. Is determined (S42).
- the CPU 1 determines that all of the reference points included in the candidate reference points PK. It is determined whether the reference point has been acquired in the process of S39 (S48). When all the reference points of the candidate reference points PK are acquired in the process of S39 (S48: YES), the CPU 1 advances the process to S50 described later. When there is a reference point that has not been acquired in the process of S39 among the reference points selected as the candidate reference point PK (S48: NO), the CPU 1 increments N by 1 (S49), and the process is the process of S39. Return to. By repeatedly executing S41, peripheral reference points a0 to a7 are added for the central reference point A1 (S41), and the CPU 1 determines that eight peripheral reference points a0 to a7 have been acquired (S42: YES).
- the CPU 1 sets the attribute information of the eight peripheral reference points and the arrangement order of the peripheral information (S43).
- the arrangement order indicates the arrangement order of the attribute information and the peripheral information of each of the eight peripheral reference points.
- the peripheral information may be the positional information itself of the peripheral reference point, or may be a calculated value calculated using the positional information of the peripheral reference point.
- the peripheral information may be one type of information or a plurality of types of information.
- the peripheral information in this example includes the ID of the peripheral reference point acquired in the process of S34, the angle calculated in the process of S36, and the distance calculated in the process of S35.
- the ID of the reference point is associated with the position information, and the peripheral information includes the ID, thereby being associated with the position information.
- the arrangement order setting method may be determined as appropriate.
- the CPU 1 in this example is on the same ridge as the center reference point PA among the plurality of peripheral reference points acquired in the processes of S39 to S42, and is the separation reference having the longest distance from the center reference point PA.
- the arrangement order is determined in the order of the peripheral reference points a0 to a7.
- the CPU 1 matches the arrangement order of the peripheral information with the arrangement order of the attribute information according to the arrangement order set in S43.
- the CPU 1 generates radiation information as part of the attribute information (S44).
- the radiation information is information indicating whether each of the peripheral reference points indicates whether the central reference point and the peripheral reference point are on the same ridge.
- the notation method of radiation information may be set as appropriate. In FIG. 11, when the center reference point and the peripheral reference point are on the same ridge (the radiation component indicated by the solid line), the center reference point and the peripheral reference point are not on the same ridge (dotted line) When the radiation component (shown) is represented by 0, the radiation information represented by the binary numbers of the peripheral reference points a0 to a7 of the central reference point A1 is 1, 0, 0, 0, 1, 0, 0, 0, 0 0 .
- the radiation information associated with the peripheral reference point on a one-to-one basis is represented by an 8-digit binary number arranged according to the arrangement order determined in the process of S43.
- binary radiation information 10001000 is associated with the center reference point A1.
- the notation method of radiation information may be changed as appropriate, such as 0x88 in hexadecimal.
- Circulation information connects two peripheral reference points that are the end points of two adjacent radiations out of a predetermined number of radiations when a central reference point and each of a predetermined number of peripheral reference points are connected This is information indicating each peripheral reference point that is a starting point for the peripheral line, indicating whether the peripheral line is on the same ridge. Which of the end points for the peripheral line is set as a starting point may be set as appropriate. In this example, when the center reference point is the center, the peripheral reference point in the counterclockwise direction in the extending direction of the peripheral line is set as the starting point. The notation method of the circulation information may be set as appropriate. In FIG.
- the circulation information represented by the binary numbers of the peripheral reference points a0 to a7 of the center reference point A1 is 0, 1, 1, 0, 0, 1, 1, 0, respectively.
- the circulation information associated with the peripheral reference point on a one-to-one basis is represented by an 8-digit binary number arranged according to the arrangement order determined in the process of S43, similarly to the radiation information.
- the binary reference information 01100110 is associated with the center reference point A1.
- the notation method of the circulation information may be changed as appropriate, such as 0x66 in hexadecimal.
- CPU 1 generates classification information as part of the attribute information (S46). For example, the CPU 1 generates classification information based on the radiation information generated in S44 and the circulation information generated in S45. More specifically, the CPU 1 is based on the number of 1 in the radiation information generated in S44 and the number of 1 in the circulation information generated in S45, for example, as in the first embodiment.
- the classification information of the center reference point A1 is generated as “0x24”.
- the CPU 1 generates sweat hole related information for the center reference point PA selected in the process of S32 and stores it in the RAM 3 (S47).
- sweat hole related information 50 is generated and stored in the RAM 3 as shown in FIG.
- the sweat hole related information 50 of the reference point A1 includes an ID 51, attribute information 52, and peripheral information 53 of the reference point A1.
- the peripheral information includes an ID, an angle, and a distance for each peripheral reference point.
- the ID is associated with the position information.
- the attribute information and the peripheral information are arranged according to the arrangement order set in S43. Among the attribute information, the classification information that is not associated one-to-one with the peripheral reference point does not have an arrangement according to the arrangement order.
- CPU 1 determines whether or not all the reference points determined in the process of S23 are selected as the central reference points in the process of S32 (S50). When there is an unselected reference point (S50: NO), the CPU 1 returns the process to S32. When all the reference points are selected as the center reference points (S50: YES), the CPU 1 ends the sweat pore related information generation processing as described above, and returns the processing to the image analysis processing of FIG. After the process of S25, the CPU 1 stores the sweat hole-related information generated in the process of S47 in the RAM 3 as authentication information (S28). The CPU 1 ends the image analysis processing as described above, and returns the processing to the authentication information processing in FIG.
- sweat hole related information is generated for a plurality of reference points including the reference points A1 to A10 shown in FIG.
- the CPU 1 generates and stores authentication information 70 including sweat hole related information 50 regarding each of the plurality of reference points.
- the CPU 1 determines whether or not authentication information including sweat hole related information is acquired in S11 (S12). When the authentication information has not been acquired (S12: NO), the CPU 1 issues an error notification (S16). For example, the CPU 1 displays an error message on the display unit 6.
- the authentication information is acquired (S12: YES)
- Information indicating whether to register is included in the start instruction, for example. In the specific example, it is determined that registration is performed (S13: YES), and the CPU 1 stores the authentication information acquired in S11 in the DB 28 of the flash memory 4 (S14).
- CPU1 When not registering in DB28 (S13: NO), CPU1 performs the collation process which makes the authentication information acquired by S11 the authentication information for collation used as collation object (S15). Next to any one of S14, S15, and S16, the CPU 1 ends the authentication information processing.
- Authentication information including sweat hole related information for a plurality of center reference points including the center reference points A1 to A10 extracted from the image 45 in FIG. A case will be described in which authentication information including sweat hole-related information for a plurality of reference points including the center reference points B1 to B9 is generated from the image 65 of FIG. 15 as an image for verification.
- S11 is executed in the same manner as the authentication information processing at the time of registration.
- authentication information including sweat pore related information for a plurality of reference points including the center reference points B1 to B9 is generated from the image 65.
- S12 of the authentication information processing in FIG. 8 it is determined that the authentication information has been acquired (S12: YES), and it is determined not to be registered based on the start instruction (S13: NO).
- CPU1 performs a collation process (S15).
- the matching process the CPU 1 compares the matching reference point determined to match the attribute information with the registration reference point, the matching authentication information used for calculating the similarity, and the registration information. The correspondence with the authentication information is determined, and the similarity of the authentication information is calculated.
- the CPU 1 executes a pair candidate extraction process (S200).
- the CPU 1 matches the attribute information associated with the reference point among the verification authentication information generated in the process of S11 and the registration authentication information stored in the DB 28.
- a pair of a reference point (sweat-related information) for registration and a reference point (sweat-related information) for registration becomes a candidate for comparison of the correspondence between the authentication information for verification and the authentication information for registration Extract as a pair candidate.
- CPU1 calculates the score of a pair candidate based on sweat hole related information.
- the CPU 1 initializes a list stored in the RAM 3 for extracting pair candidates (S211).
- the CPU 1 selects one reference point (center reference point) that has not yet been selected in S212 among the reference points in which the sweat hole related information included in the verification authentication information generated in the process of S11 is generated. (S212).
- the CPU 1 selects the reference point B1 from the reference points where the sweat hole-related information included in the verification authentication information is generated.
- the CPU 1 selects one reference point (center reference point) that has not yet been selected in the process of S213 among the reference points where the sweat pore related information included in the authentication information for registration registered in the DB 28 is generated. (S213).
- the CPU 1 selects the reference point A1 from the reference points where the sweat hole-related information included in the authentication information for registration is generated.
- the CPU 1 determines whether the classification information of the associated attribute information matches between the reference point for matching selected in the process of S212 and the reference point for registration selected in the process of S213 ( S214).
- the classification information of the reference point B1 is 0x14, which is different from 0x24 which is the classification information of the reference point A1 (S214: NO).
- the CPU 1 does not extract the combination of the reference point B1 selected in S212 and the reference point A1 selected in S213 as a pair candidate, and all reference points for registration are selected in the process of S213. (S224). If there is a registration reference point that has not been selected (S224: NO), the CPU 1 returns the process to S213.
- the classification information of the reference point B1 is 0x14, which matches the classification information of the reference point A2 0x14 (S214: YES).
- the CPU 1 sets the variable N to 0 (S215).
- the variable N is a variable for sequentially comparing attribute information (radiation information and circulation information in this example) corresponding to the peripheral reference points on a one-to-one basis in consideration of image shrinkage and rotation.
- the CPU 1 determines whether the radiation information of the associated attribute information matches between the reference point for matching selected in the process of S212 and the reference point for registration selected in the process of S213 ( S216).
- the CPU 1 compares the radiation information of the reference point for verification with the radiation information of the registration reference point corresponding to the variable N. As shown in FIG. 19, when the variable N is 0 for the reference point A2, the radiation information corresponding to the peripheral reference points b0 to b7 is compared with the radiation information corresponding to the peripheral reference points a0 to a7, respectively. When the variable N is 1, the radiation information corresponding to the peripheral reference points b0 to b7 is compared with the radiation information corresponding to the peripheral reference points a1 to a7 and a0, respectively. That is, the CPU 1 changes the reference start position of the registration reference point attribute information according to the variable N.
- the reference point bm for verification (m is an integer from 0 to 7) is compared with the reference point a for registration (MOD (N + m, 8)).
- MOD (N + m, 8) is the remainder when the sum of N and m is divided by 8.
- the radiation information of the reference point B1 is 10000000, which coincides with the radiation information of the reference point A2 when the variable N is 0 (S216: YES).
- the CPU 1 determines whether the circulation information in the associated attribute information matches between the reference point for comparison selected in the process of S212 and the reference point for registration selected in the process of S213. Judgment is made (S217).
- the processing of S217 also compares the reference information of the reference point for verification with the reference information of the reference point for registration corresponding to the variable N.
- the circulation information of the reference point B1 is 01100110, which is consistent with 01100110, which is the circulation information of the reference point A2 when the variable N is 0 (S217: YES).
- the CPU 1 uses the score representing the similarity between the reference point for matching selected in the process of S212 and the reference point for registration corresponding to the variable N selected in the process of S213 as the sweat hole related information. It is calculated from the peripheral information within (S218).
- the peripheral information of the sweat hole related information in this example includes an ID, an angle, and a distance associated with the position information.
- the CPU 1 of this example uses, for example, the reference amount for registration and the reference point for verification to calculate the score using the comparison amounts of the eight sets of angles and the distance comparison amounts of the peripheral reference points currently being compared. calculate.
- the variable N is 0, as shown in FIG. 20, the score matches the center reference point between the registration reference point A2 and the reference reference point B1, and the rotation angle is considered.
- 8 represents the amount of deviation between the angles and distances of the eight peripheral reference points when the arrangement of the eight peripheral reference points is compared in a state in which one set of radiation components serving as a reference is superimposed.
- a set of radiation components serving as a reference based on N is a combination of radiation components whose end points are a reference point b0 for verification and a reference point a (MOD (N, 8)) for registration.
- the eight sets of peripheral reference points are reference points of combinations indicated by double arrows in FIG. 19, and a reference point bm for matching and a reference point a (MOD ( N + m, 8)).
- the rotation angle is an angle calculated based on sweat hole related information that indicates how many times the verification image is assumed to be rotated with respect to the registration image.
- the rotation angle in this example is a positive angle in the clockwise direction in the image coordinate system.
- the comparison amount is a value calculated by comparing the peripheral information in the sweat pore related information between the reference point for verification and the reference point for registration.
- the comparison amount of the eight sets of angles May be represented by, for example, the sum of squares of the difference in rotation angle calculated by Expression (1) for each of integers from 0 to 7.
- Difference in rotation angle Bnm ⁇ An (MOD (N + m, 8)) + AnN ⁇ Bn0 Formula (1)
- the comparison amount of the eight sets of distances For example, m may be represented by the sum of the ratios of the distance differences calculated by Expression (2) for each of the integers from 0 to 7.
- Score (Comparison amount maximum value ⁇ (Distance comparison amount ⁇ Constant + Angle comparison amount)) / Comparison amount maximum value ⁇ Score maximum value Equation (3)
- the constant is a value that is appropriately set to adjust the magnitude of the distance comparison amount with respect to the angle comparison amount, and is 100, for example.
- the maximum comparison amount is the maximum value that is allowed as the sum of the value obtained by multiplying the distance comparison amount by a constant and the angle comparison amount. When the sum of the distance comparison amount multiplied by a constant and the angle comparison amount is greater than the comparison amount maximum value, a value obtained by multiplying the distance comparison amount by a constant in equation (3); The maximum comparison amount is set to the sum of the angle comparison amount.
- the maximum score value is the maximum value that can be taken by the score, and is 100, for example.
- the rotation angle between the reference point B1 for verification and the reference point A2 for registration is calculated as 41 degrees and stored in the list.
- N determines whether N is greater than 0 (S219). When N is larger than 0 (S219: YES), the CPU 1 determines whether the score calculated in the process of S218 is larger than the score stored in the list (S220). When N is not larger than 0 (S219: NO), or when the score is larger than the stored score (S220: YES), the CPU 1 displays the score and rotation angle calculated in S218 in the list in the RAM 3. (S221). By this processing, the value having the highest attribute information and the highest score is stored in the list between the reference point for matching and the reference point for registration.
- the CPU 1 increments the variable N by 1 (S222).
- the CPU 1 determines whether N is smaller than 8 (S223).
- the threshold value 8 of S223 is the same as the number of peripheral reference points, and is set for comparing attribute information with all combinations considering the influence of image rotation.
- N is smaller than 8 (S223: YES)
- the CPU 1 returns the process to S216.
- N is not smaller than 8 (S223: NO)
- the CPU 1 determines whether all reference points included in the authentication information stored in the DB 28 have been selected in the process of S213 (S224).
- the CPU 1 If there is a reference point for verification that has not been selected (S225: NO), the CPU 1 returns the process to S212.
- the CPU 1 ends the pair candidate extraction process as described above, and returns the process to the collation process of FIG.
- the pair candidate extraction process extracts a score for each of the registration reference point (sweat related information) extracted as a pair candidate and the matching reference point (sweat related information).
- the rotation angle is acquired.
- the upper value of each combination of the registration reference point and the reference point for matching is the score, and the lower value indicates the rotation angle.
- a combination in which the score and the rotation angle are blank is a combination that has not been extracted as a pair candidate, and is a combination in which attribute information does not match.
- a plurality of reference points for registration are extracted as pair candidates for one reference point for reference (sweat related information).
- registration reference points A1, A3, A4, A7, A8, and A10 are extracted as pair candidates for the reference point B3 for verification.
- 36 sets are pair candidates. Extracted.
- CPU1 determines the correspondence between the skin information for verification and the skin information for registration by comparing the peripheral information in the sweat pore related information for the pair candidates extracted in the process of S200 after S200. Specifically, the CPU 1 extracts a pair of related information extracted as a pair candidate and a combination (pair, correspondence) of the related information for registration, the related information for verification, and the related information for registration A plurality of pieces of image information including at least one of the rotation amount and the movement amount between the comparison image calculated from the above and the registration image are acquired (S201).
- the image information of this example is stored in the process of S221, and is a rotation angle indicating the amount of rotation between the image for verification and the image for registration.
- CPU1 acquires the combination of a pair candidate shown in FIG.
- the CPU 1 narrows down the pair candidates by comparing the image information of the combination of the acquired plurality of sets of related information for collation and the related information for registration (S202).
- the CPU 1 narrows down the pair candidates whose image information falls within a predetermined range set based on the representative value of the acquired image information of a plurality of pairs of pairs.
- the representative value may be a value representing the characteristics of a plurality of image information, and may be, for example, an average value, a mode value, a median value, or the like of the image information.
- the CPU 1 of this example classifies the rotation angles of the pair candidates into a predetermined range of angles, extracts a pair candidate having a rotation angle that falls within a predetermined range plus or minus from the representative value of the rotation angle range that appears most frequently, Narrow down candidates.
- the predetermined range is preferably a range of 1 degree (360 resolution) to 20 degrees (18 resolution).
- the predetermined angle is appropriately determined in consideration of a predetermined range, preferably includes at least three resolutions, and is an angle that is a representative value plus or minus 15 degrees or more and 60 degrees or less. In a specific example, as shown in FIG.
- the CPU 1 sets the rotation angle of the pair candidate every 10 degrees (for example, a range where the representative value is expressed as 5 degrees and less than 10 degrees, and the representative value is 15 degrees. In the range of 10 degrees or more and less than 20 degrees, etc. In this case, as shown in FIG. 22, the range in which the representative value is 45 degrees has the highest frequency.
- the CPU 1 sets the representative value in the range with the highest frequency as the representative value of the image information, and sets the candidate candidates that fall within the range of 30 degrees to 60 degrees with the representative value of 45 degrees plus or minus 15 degrees. Extract.
- the rotation angle is 44 degrees, which is a value in the range of 30 degrees to 60 degrees, and thus is extracted in the process of S202.
- the rotation angle is 290 degrees, which is not a value in the range of 30 degrees to 60 degrees, and therefore is not extracted in the process of S202.
- the combination including the combination of A1 and B4 indicated by dot shading in FIG. 21 is a combination that has not been extracted in the process of S202.
- the CPU 1 takes the fact that the distribution may vary when the rotation angle of the pair candidate is classified into a predetermined range of angles, and takes the moving average of the frequencies to identify the range with the highest frequency. May be.
- CPU 1 extracts the pair candidate having the highest score when the reference point for matching and the reference point for registration do not have a one-to-one correspondence among the pair candidates narrowed down in S202.
- the reference point for registration and the reference point for registration are narrowed down so as to correspond one-to-one (S203).
- the CPU 1 in this example narrows down the pair candidates shown in FIG. 21 to the pair candidates that are shaded.
- the CPU 1 compares at least one of the length and the angle of the line segment connecting each of the center reference points of a plurality of pair candidates arbitrarily selected from the pair candidates narrowed down in S202 and S203.
- the pair candidates are further narrowed down (S204). Conditions for selecting a plurality of pairs of candidates may be changed as appropriate.
- the CPU 1 in this example selects two pairs of arbitrary pair candidates from the pair candidates narrowed down in the process of S203, and the pair candidate is determined based on the positional relationship of the line segment connecting the reference points of the two selected pair candidates. Further narrow down.
- each of the angle of the line segment connecting the reference points of the two selected pair candidates and the length of the line segment satisfies a predetermined condition is extracted, and the pair candidates are further narrowed down.
- the pair candidate of the reference points A5 and B2 and the reference points A1 and B3 are selected as the two pairs of candidates, the line segment connecting the reference points A5 and A1, and the reference With respect to the line segment connecting the points B2 and B3, it is determined whether the angle regarding the two line segments and the length of the line segment satisfy a predetermined condition.
- the condition regarding the angle is a condition in which the difference between the angle of the line connecting the reference points A5 and A1 and the angle of the line connecting the reference points B2 and B3 falls within a predetermined range.
- the predetermined range is, for example, a range of plus or minus 5 degrees of the representative value of the angle.
- the representative value is, for example, the average value of the rotation angles of the pair candidates narrowed down in the process of S203.
- the predetermined range in S204 is narrower than the range in S202.
- the condition regarding the length is a condition in which the length d1 of the line segment connecting the reference points A5 and A1 and the length d2 of the line segment connecting the reference points B2 and B3 satisfy Expression (4).
- the CPU 1 narrows down the pair candidates based on the condition regarding the angle and the condition regarding the line segment.
- the CPU 1 in this example selects two pairs of candidates, compares the lengths and angles of the line segments connecting the center reference points of the pair candidates, and repeats the process of narrowing down the pair candidates until a predetermined condition is satisfied.
- the predetermined condition is a condition in which, for example, all combination pair candidates narrowed down by the processes of S202 and S203 are selected, and it is determined whether a condition regarding an angle and a condition regarding a line segment are satisfied.
- the CPU1 determines the pair candidate narrowed down by the process of S202 to S204 as a pair (S205). In addition to S202 to S204, the CPU 1 may determine pairs by narrowing down pair candidates under other conditions.
- the CPU 1 uses the correspondence between the verification authentication information (reference point) and the registration authentication information (reference point), which are narrowed down in the processing of S202 to S204 and determined as a pair in the processing of S205.
- the similarity between the authentication information and the registration authentication information is calculated (S206).
- the CPU 1 calculates the score SC using the sum of the pair scores determined in S205. For example, the CPU 1 uses the sum of the pair scores determined in S205 as the score SC.
- the CPU 1 may calculate the score SC by substituting the sum of the similarities into a predetermined formula.
- the score SC indicates, for example, that the larger the value is, the more similar the verification authentication information and the registration authentication information are compared to the case where the value is small.
- the CPU 1 determines whether or not the similarity (score SC) calculated in S206 is larger than a threshold (S207). When the similarity is larger than the threshold (S207: YES), the CPU 1 sets success in the authentication result of the skin authentication (S208). When the similarity is not greater than the threshold (S207: NO), the CPU 1 sets failure in the authentication result of the skin authentication (S209).
- the CPU 1 may make a notification such as displaying the authentication result on the display unit 6 as necessary in the processes of S208 and S209.
- Condition 1 is a condition for performing skin authentication using a known minutiae method.
- Condition 2 is a condition for performing skin authentication using sweat pore-related information.
- Condition 3 is a condition for performing skin authentication by using the minutiae method and sweat pore related information together. The test results of conditions 1 to 3 are shown as results 35 to 37 in FIG. 24, respectively.
- the condition 11 is a condition that does not narrow down the pair candidates extracted in the process of S200 of FIG. 16 (condition that does not execute the processes of S202 to S204).
- Condition 12 is a condition for narrowing down the pair candidates extracted in the process of S200 using image information (rotation angle) (a condition in which the process of S202 is performed and the processes of S203 and S204 are not performed).
- Condition 13 is a condition for narrowing down the pair candidates extracted in the process of S200 using image information (rotation angle) and score (conditions in which the processes of S202 and S203 are performed and the process of S204 is not performed).
- Condition 14 is a condition for narrowing down the pair candidates extracted in the process of S200 using the positional relationship of a plurality of pairs of candidates (conditions for performing the process of S204 without performing the processes of S202 and S203).
- Condition 15 is a condition for narrowing down the pair candidates extracted in the process of S200 by the processes of S202 to S204.
- the results of conditions 11 to 15 are shown as results 55 to 59 in FIG. 25, respectively.
- the condition 12 (result 56) is superior to the condition 11 (result 55) in the authentication performance.
- condition 13 was further superior in authentication performance. Therefore, by implementing S202 and narrowing down the pair candidates with the image information, the authentication performance can be improved as compared with the case where the pair candidates are not narrowed down with the image information, and by executing the processing of S203 in addition to S202.
- Condition 14 (result 58) also had better authentication performance than condition 11 (result 55). From this, it was confirmed that by narrowing down the pair candidates in the process of S204, the authentication performance can be improved compared to the case where the pair candidates are not narrowed down in S204. Furthermore, compared to conditions 12 to 14, condition 15 was superior in authentication performance. As a result, it was confirmed that by performing the processing of S204 in addition to the processing of S202 and S203, the authentication performance can be improved as compared with the case where both are executed alone.
- the processing times of conditions 13 to 15 were 5.4 milliseconds, 18.5 milliseconds, and 10.9 milliseconds, respectively.
- the treatment time for condition 12 was about 1/100 of condition 14.
- the processing time under condition 14 required slightly more than three times that of condition 13.
- the treatment time for condition 15 was approximately twice the treatment time for condition 13, but less than condition 14. From this, it was confirmed that both the speeding up of the authentication processing and the improvement of the authentication performance were realized by narrowing down the pair candidates in order from S202 to S204.
- the CPU 1 is an example of the processor of the present invention.
- the ROM 2, RAM 3 and flash memory 4 are examples of the storage device of the present invention.
- the process of S21 in FIG. 9 is an example of an image acquisition step, and the CPU 1 that executes the process of S21 is an example of an image acquisition unit.
- the process of S23 is an example of a reference point determination step of the present invention, and the CPU 1 that executes the process of S23 is a reference point determination unit of the present invention.
- the processing from S39 to S42 is an example of the peripheral reference point extracting step of the present invention, and the CPU 1 that executes the processing of S39 to S42 is an example of the extracting means of the present invention.
- the process of S47 is an example of the information generation step of the present invention, and the CPU 1 that executes the process of S47 is an example of the information generation unit of the present invention.
- the process of S28 is an example of the storage control step of the present invention, and the CPU 1 that executes the process of S28 is an example of the storage control means of the present invention.
- the processing from S214 to S217 is an example of the pair candidate extraction step of the present invention.
- the processing from S202 to S205 is an example of a correspondence determination step of the present invention.
- the process of S206 is an example of the similarity calculation step of the present invention.
- the apparatus 10 can generate authentication information including sweat pore related information and store it in the RAM 3 or the DB 28 (S28, S14).
- the placement of sweat pores on the skin ridge is as unique as fingerprints and voiceprints and is said to remain unchanged throughout life. There is a possibility that a plurality of sweat holes can be acquired even when the size of the image representing the skin is smaller than the conventional size and the branch points and end points of the ridges are not included in the image.
- the reference point represents a sweat hole on the ridge
- the attribute information represents a feature of the arrangement of each of the predetermined peripheral reference points on the image.
- the attribute information is information that represents the feature of the arrangement of a predetermined number of sweat holes around the target sweat hole, and can be said to be information that emphasizes the feature portion of the biological information represented by the image.
- the attribute information includes, for example, verification authentication information used for calculating the similarity of authentication information, reference points for verification used for processing to determine correspondence between the authentication information for registration, and reference points for registration. Can be suitably used for the process of extracting. Therefore, the apparatus 10 can generate the sweat hole related information that is information used for the skin authentication and contributes to the improvement of the authentication speed as compared with the prior art by executing the authentication information processing.
- the apparatus 10 determines the center of gravity of the sweat hole in the image as a reference point, and acquires position information of the reference point. Therefore, the apparatus 10 can determine the center of gravity of the sweat hole representing the feature of the shape and size of the sweat hole as a reference point and generate sweat hole related information.
- the apparatus 10 forms a line segment connecting the already selected peripheral reference point and the center reference point, and a line segment connecting the reference point that is a candidate for the peripheral reference point and the center reference point.
- Peripheral reference points are extracted under the condition that a predetermined number of reference points whose corner angles are equal to or greater than a predetermined angle are selected in order of increasing distance from the central reference point. Therefore, the apparatus 10 suppresses that only a reference point representing a sweat hole on the same ridge as the center reference point is extracted as a peripheral reference point, and includes sweat information including attribute information that suitably represents features around the center reference point. Hole-related information can be generated.
- the attribute information includes classification information, and the classification information includes first information indicating the number of peripheral reference points on the same ridge as the central reference point among predetermined peripheral reference points.
- the classification information further includes a predetermined reference line segment when the central reference point and each of the predetermined peripheral reference points are connected as a radiation component, and the peripheral reference points on two adjacent radiation components as the central reference. It includes classification information including second information indicating the number of peripheral lines on the ridge when each of a predetermined number of line segments sequentially connected around the point is a peripheral line.
- the classification information suitably represents the difference in the arrangement of the peripheral reference points from the two viewpoints of the first information and the second information.
- the apparatus 10 uses the verification authentication information used for calculating the similarity of the authentication information, the verification reference point used for processing for determining the correspondence between the registration authentication information, and the registration reference point.
- Classification information that can contribute to speeding up the extraction process can be generated.
- the CPU 1 can avoid the process of determining attribute information in consideration of contraction and rotation of the image for combinations that are unlikely to match other attribute information, and can shorten the processing time.
- the attribute information includes radiation information that is information indicating whether each of the peripheral reference points indicates whether the central reference point and the peripheral reference point are on the same ridge (S44).
- the apparatus 10 uses the verification authentication information used for calculating the similarity of the authentication information, the verification reference point used for processing for determining the correspondence between the registration authentication information, and the registration reference point. Radiation information that can contribute to speeding up the extraction process can be generated. Since the CPU 1 in this example determines whether the radiation information matches for all combinations that consider the contraction and rotation of the image at the peripheral reference point, it determines whether the radiation information matches in consideration of the contraction and rotation of the image. I can judge.
- Attribute information includes lap information (S45).
- Circulation information includes a central reference point and each of a predetermined number of peripheral reference points as a radiation component and a peripheral reference point on two adjacent radiation components as a central reference point.
- Circulation information is information from a different viewpoint from radiation information, and can be said to be information that takes into account the arrangement of the central reference point and the arrangement of other peripheral reference points.
- the apparatus 10 uses the verification authentication information used for calculating the similarity of the authentication information, the verification reference point used for processing for determining the correspondence between the registration authentication information, and the registration reference point. Circulation information that can contribute to speeding up the extraction process can be generated.
- the apparatus 10 For each of the plurality of reference points, the apparatus 10 is based on the position information of a predetermined number of peripheral reference points in addition to the position information of the center reference point (specifically, the ID associated with the position information) and the attribute information.
- the sweat hole related information including the peripheral information which is information is generated (S47).
- the device 10 can generate sweat hole related information including peripheral information that can be used for processing for determining the correspondence between the authentication information for verification used for calculating the similarity of skin authentication and the authentication information for registration.
- the apparatus 10 calculates the distance from the center reference point for each of the predetermined peripheral reference points (S35).
- the device 10 generates sweat hole related information including the distance as the peripheral information (S47).
- the apparatus 10 can generate sweat hole related information including a distance that can be used for processing for determining the correspondence between the authentication information for verification used for calculating the similarity of skin authentication and the authentication information for registration.
- the apparatus 10 can shorten the time required for skin authentication as compared with the case where the distance is calculated every time skin authentication is performed.
- the apparatus 10 calculates an angle with respect to a predetermined direction of a line segment connecting the central reference point and the peripheral reference point for each of the predetermined peripheral reference points (S36).
- the apparatus 10 generates sweat hole related information including an angle as peripheral information (S47).
- the apparatus 10 can generate sweat hole related information including an angle that can be used for processing for determining the correspondence between the authentication information for verification used for calculating the similarity of the authentication information and the authentication information for registration.
- the apparatus 10 can shorten the time required for skin authentication as compared with the case where the angle is calculated every time skin authentication is performed.
- the arrangement order of the peripheral information and the arrangement order of the attribute information are made coincident, and among the predetermined peripheral reference points, they are on the same ridge as the central reference point, and the distance from the central reference point is
- the arrangement order of the longest separation reference point is set to No. 1
- the arrangement order of the second and subsequent arrangements is set around a predetermined direction from the separation reference point centered on the center reference point based on the arrangement on the image (S43).
- the distance between sweat holes on the same ridge is less susceptible to image distortion than the distance between sweat holes on different ridges. For this reason, the apparatus 10 can generate the sweat hole related information in which the arrangement order is set based on the combination of the center reference point and the peripheral reference point considered to have the least influence of image distortion.
- the apparatus 10 includes a matching reference point that matches the attribute information associated with the reference point among the matching authentication information generated in the process of S47 and the registration authentication information stored in the DB 28.
- a pair with a reference point for registration is extracted as a pair candidate for which the correspondence between the authentication information for verification and the authentication information for registration is compared (S214 to S217).
- the device 10 determines the correspondence between the verification authentication information and the registration authentication information using the pair candidates (S202 to S205).
- the apparatus 10 is shorter in time than the case where the correspondence between the authentication information for verification used for calculating the similarity and the authentication information for registration is determined by comparing all the reference points included in the authentication information. Can be determined.
- the apparatus 10 of this example extracts pair candidates step by step by sequentially comparing each of a plurality of attribute information. For this reason, compared with the case where one attribute information is used, a pair candidate can be narrowed down efficiently from many viewpoints.
- the device 10 determines whether the attribute information associated with the reference point for matching and the reference point for registration match in consideration of the effects of rotation and contraction of the skin information at the time of acquisition. it can.
- the peripheral reference point with respect to the central reference point for matching is determined based on whether or not the circulation information associated with the acquired reference point for matching matches the circulating information associated with the reference point for registration.
- the device 10 determines whether or not the attribute information associated with the reference point for matching and the reference point for registration match based on the rotation information in consideration of the effect of rotation and contraction of the skin information at the time of acquisition. it can.
- the device 10 further compares the peripheral information in the sweat hole related information with respect to the extracted pair candidates and determines the correspondence between the verification authentication information and the registration authentication information (S202 to S205).
- the device 10 can compare the peripheral information of the pair candidates and determine the correspondence between the verification authentication information and the registration authentication information.
- the apparatus 10 of this example determines the correspondence based on the score calculated based on the peripheral information and the rotation angle. For this reason, the apparatus 10 can determine a response
- the apparatus 10 uses the correspondence between the verification authentication information determined in the processing of S202 to S205, the registration authentication information stored in the DB 28, and the verification authentication information and the registration authentication information.
- the similarity is calculated (S206).
- the apparatus 10 can execute processing from generation of authentication information to similarity calculation in a relatively short time.
- the authentication information processing program and authentication information processing apparatus of the present invention are not limited to the above-described embodiment, and various modifications may be made without departing from the scope of the present invention. For example, the following modifications (A) to (C) may be added as appropriate.
- the configuration of the apparatus 10 may be changed as appropriate.
- the device 10 is not limited to a smartphone, and may be a mobile device such as a notebook PC, a tablet PC, and a mobile phone, or an automatic teller machine (ATM) and entrance / exit management. It may be a device such as a device.
- the biological information acquisition device 8 may be provided separately from the device 10. In that case, the biometric information acquisition apparatus 8 and the apparatus 10 may be connected by a connection cable, or may be connected wirelessly such as Bluetooth (registered trademark) and NFC (Near Field Communication).
- the detection method of the biological information acquisition device 8 may be, for example, an electric field method, a pressure method, or an optical method.
- the biological information acquisition device 8 is not limited to a plane type, and may be a linear type.
- the size, color information, and resolution of the image generated by the biometric information acquisition device 8 may be appropriately changed as long as sweat holes can be extracted. Therefore, for example, the color information may be information corresponding to a color image in addition to information corresponding to a black and white image.
- the authentication information processing program may be stored in the storage device of the device 10 before the device 10 executes the program. Therefore, each of the authentication information processing program acquisition method, the acquisition route, and the device that stores the authentication information processing program may be changed as appropriate.
- the information processing program executed by the processor of the device 10 may be received from another device via a cable or wireless communication and stored in a storage device such as a flash memory.
- Other devices include, for example, a PC (personal computer) and a server connected via a network.
- the storage device may be a non-temporary storage medium such as HDD and SSD in addition to the ROM 2 and the flash memory 4, and may be any storage medium that can retain information regardless of the information storage period. Good.
- the non-transitory storage medium may not include a temporary storage medium (for example, a signal to be transmitted).
- Each step of the authentication information processing is not limited to the example executed by the CPU 1, and a part or all of the steps may be executed by another electronic device (for example, ASIC).
- Each step of the above process may be distributedly processed by a plurality of electronic devices (for example, a plurality of CPUs).
- Each step of the authentication information processing in the above embodiment can be changed in order, omitted, or added as necessary.
- an operating system (OS) or the like operating on the device 10 performs part or all of the actual processing based on a command from the CPU 1 of the device 10, and the functions of the above-described embodiments are realized by the processing.
- OS operating system
- the following changes (C-1) to (C-4) may be appropriately added to the authentication information processing.
- Pre-processing may be appropriately performed on the image acquired in S11.
- a filtering process for removing high frequency components of an image as noise may be executed. By executing the filtering process, the change in shading of the edge portion of the image is moderated.
- Any of a known low-pass filter, Gaussian filter, moving average filter, median filter, and averaging filter may be used as a filter used in the filtering process.
- a filtering process for extracting only a specific frequency band component may be performed on the image acquired in S11.
- a specific frequency band a band including the period of the concave and convex portions of the fingerprint may be selected.
- a known band-pass filter may be used as a filter used for the filtering process.
- the reference point may be a point representing a sweat hole and may not be the area center of gravity of the sweat hole.
- the sweat hole related information may not be generated for all the reference points determined from the image.
- the peripheral reference point extraction conditions predetermined angle, predetermined number, predetermined condition, etc.
- the sweat hole related information may include one or more types of attribute information.
- the attribute information may be information that is not one-to-one correspondence with the peripheral reference point like the classification information, or may be one-to-one correspondence with the peripheral reference point like the radiation information and the circulation information. It may be information.
- the classification information may include only one of the first information and the second information, or may include other information in addition to at least one of the first information and the second information. When the classification information includes a plurality of pieces of information including the first information and the second information, the arrangement of each information may be changed as appropriate.
- the sweat hole related information may not include the peripheral information.
- the method for setting the position information may be changed as appropriate.
- the peripheral information may be information based on position information, may include any of position information, an angle, and a distance, or other information calculated based on position information. May be included.
- the method for determining the arrangement order may be changed as appropriate.
- the arrangement order may be an acquisition order of the peripheral reference points.
- the order of the processes from S43 to S46 may be changed as appropriate.
- the order of the process of S44 and the process of S45 may be interchanged, or may be executed in parallel.
- the process of S43 may be executed after the process of S44 and the process of S45.
- the authentication information including the generated sweat pore related information may not necessarily be used for the process of calculating the similarity used for skin authentication.
- the method for determining the correspondence of the processing from S202 to S205 may be appropriately changed.
- the apparatus 10 may determine the correspondence by comparing the peripheral information, or may determine the correspondence based on the arrangement with other reference points other than the peripheral reference points.
- the apparatus 10 determines correspondence between the two reference points extracted as pair candidates by comparing other information such as known frequency information associated with each of the reference points (see, for example, Japanese Patent Application Laid-Open No. 2017-010419). May be.
- Skin authentication may be performed by a combination of perforation related information and known authentication information.
- the final determination may be executed by combining a matching result obtained by a known minutiae method and a matching result obtained using the authentication method of the present invention.
- the verification method may be set automatically or by the user from a plurality of types of verification methods in consideration of processing time, authentication accuracy, and the like.
- the final determination may be performed by combining verification results using an authentication method using known frequency information.
- the frequency information in this case may be information indicating the color change around the reference point.
- the frequency component is not limited to a one-dimensional group delay spectrum.
- frequency component such as an LPC spectrum, a group delay spectrum, an LPC cepstrum, a cepstrum, an autocorrelation function, and a cross correlation function may be used as the frequency component.
- the frequency information may be stored in association with the reference point.
- the authentication information program includes a sample information acquisition step for acquiring sample information, which is information representing a change in color information around the reference point determined in the reference point determination step, and a sample acquired in the sample information acquisition step. It further includes an instruction for executing a frequency information calculation step of calculating information associating the frequency component of the information with the position information as frequency information, and related to the sweat pores generated in the information generation step in the storage control step.
- the information may be stored in the storage device as authentication information in association with the frequency information acquired in the frequency information calculation step.
- the sample information acquisition step and the frequency information calculation step may be executed, for example, between the processes of S25 and S28 in FIG.
- the similarity calculation step based on the correspondence determined in the correspondence determination step, the similarity that is the similarity between the matching sweat hole related information and frequency information and the registration sweat hole related information and frequency information is obtained. It may be calculated.
- the method for determining the correspondence (pair) from the pair candidates may be determined as appropriate.
- the CPU 1 in this example uses the score and rotation angle calculated based on the peripheral information to select combinations extracted as pair candidates, and each combination of the registration reference point and the reference point for matching is a pair. Decide to be one. More specifically, for example, a combination having a high score may be preferentially determined as a pair. A combination whose rotation angle is significantly different from other pair candidates may not be determined as a pair.
- the score calculation method and the similarity calculation method may be appropriately changed.
- the pair candidates may be narrowed down based on at least one of the rotation angle and the movement amount of the collation image with respect to the registration image. Part or all of the processing from S200 to S206 may be omitted as appropriate.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Security & Cryptography (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Pathology (AREA)
- Biophysics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Collating Specific Patterns (AREA)
Abstract
Description
図2から図7を参照して、第一実施形態の装置10において実行される認証情報記憶処理を説明する。図2に示すように、装置10は、生体情報取得装置8、画像取得部21、基準点決定部22、抽出部23、情報生成部24、登録部26、照合部27、及びDB28を有し、それぞれの機能ブロックに対応する処理を、CPU1(図1参照)によって実行する。
1.登録時の処理
図4、図5、図8から図23を参照して、第二実施形態の装置10で実行される認証情報処理について、認証情報を登録する場合を例に説明する。認証情報処理は、ユーザが開始指示を入力した場合に開始される。開始指示には、画像から取得された認証情報を登録用の認証情報としてDB28に登録するのか、それとも、取得された認証情報をDB28に登録された登録用の認証情報と照合するのかに関する指示を含む。装置10のCPU1は、認証情報処理の開始指示の入力を検知すると、フラッシュメモリ4に記憶された認証情報処理を実行する為の認証情報処理プログラムをRAM3に読み出し、認証情報処理プログラムに含まれる指示に従って、以下に説明する各ステップの処理を実行する。本例では、基準点を抽出する要件(例えば、画像の鮮明さ)を満たす皮膚情報が取得されるまで、再入力を促すフィードバック処理が実行される。認証情報処理で取得される画像は、画像からアルゴリズムを用いて認証情報を抽出する要件を満たす。処理の過程で取得されたり、生成されたりした情報及びデータは、適宜RAM3に記憶される。処理に必要な各種設定値は、予めフラッシュメモリ4に記憶されている。以下、ステップを「S」と略記する。
登録用の認証情報として図13の画像45から抽出された中心基準点A1からA10を含む複数の中心基準点についての汗孔関連情報を含む認証情報が用いられ、照合対象となる照合用の画像として図15の画像65から中心基準点B1からB9を含む複数の基準点についての汗孔関連情報を含む認証情報が生成された場合について説明する。照合時の認証情報処理では、登録時の認証情報処理と同様に、S11が実行される。S11の処理によって、画像65から中心基準点B1からB9を含む複数の基準点についての汗孔関連情報を含む認証情報が生成される。
回転角の差=Bnm-An(MOD(N+m,8))+AnN-Bn0 ・・・ 式(1)
照合用の基準点bmの距離を距離Dbm、登録用の基準点a(MOD(N+m,8))の距離を距離Da(MOD(N+m,8))としたとき、8組の距離の比較量は、例えば、mが0から7までの整数の各々の、式(2)で算出される距離の差の比率の和で表されてもよい。
距離の差の比率=|Dbm-Da(MOD(N+m,8))|/min(Dbm,Da(MOD(N+m,8))) ・・・ 式(2)
ただし、min(Dbm,Da(MOD(N+m,8)))は、DbmとDa(MOD(N+m,8))とのうちの小さい方の値である。角度の比較量及び距離の比較量が0である場合のスコア(スコア最大値)を100とした場合、式(3)に基づき、例えば照合用の基準点B1と、登録用の基準点A2とのスコアが、85と算出され、リストに記憶される。
スコア=(比較量最大値-(距離の比較量×定数+角度の比較量))/比較量最大値×スコア最大値 ・・・式(3)
式(3)において定数は、角度の比較量に対する、距離の比較量の大きさを調整する為に適宜設定される値であり、例えば、100である。比較量最大値は、距離の比較量に定数を乗じた値と、角度の比較量との和の許容される最大値である。距離の比較量に定数を乗じた値と、角度の比較量との和が比較量最大値よりも大きくなる場合は、式(3)の内、距離の比較量に定数を乗じた値と、角度の比較量との和に比較量最大値が設定される。スコア最大値は、スコアが取り得る値の最大値であり、例えば、100である。照合用の基準点B1と、登録用の基準点A2との回転角は、41度と算出され、リストに記憶される。
|d1-d2|×2/(d1+d2)<0.1 ・・・式(4)
汗孔関連情報を照合に用いることで、認証性能が向上するかを確認する評価試験を行った。後述の条件1から3の各々について、31指を光学式タッチセンサによって2000dpiの横480ピクセル縦800ピクセルの画像を1指あたり5~10枚取得し、1枚を登録画像、他を照合画像としてROC(Receiver Operating Characteristic)を算出し、認証精度の比較を行った。条件1は、公知のマニューシャ法を用いて皮膚認証を行う条件である。条件2は、汗孔関連情報を用いて皮膚認証を行う条件である。条件3は、マニューシャ法と汗孔関連情報とを併用して皮膚認証を行う条件である。条件1から3の試験結果を各々、図24の結果35から37で示す。図24に示すように、条件1から3を比較すると、条件1(結果35)に比べ条件2(結果36)は、認証性能が優れていた。このことから、汗孔関連情報は、既存の認証方法の認証性能を向上させることができることが確認された。条件1(結果35)に比べ条件3(結果37)は、認証性能が優れていた。このことから、汗孔関連情報は既存の認証方法と組み合わせることにより、既存の認証方法の認証性能を向上させることができることが確認された。
属性情報を用いてペア候補を抽出することで、認証速度が向上するかを確認する評価試験を行った。属性情報を用いてペア候補を抽出する条件(図17のS214、S216、S217を実行する条件)と、属性情報を用いてペア候補を抽出しない条件(図17のS214、S216、S217を実行しない条件)とで、他の条件は同じ条件とし、同じ指紋画像データベースを用いて認証情報処理を実行させた。その結果、属性情報が無しの条件では認証までに33分41秒を要し、属性情報が有りの条件では認証までに2分12秒を要した。属性情報が無しの条件と、属性情報が有りの条件とで、認証性能を示すEER(Equal Error Rate)は同じであった。以上から、装置10は、属性情報を用いてペア候補を抽出することで、認証性能を落とさずに、処理速度が約15倍高速化されたことが確認された。
画像情報を用いてペア候補を絞り込むことで、認証性能が向上するかを確認する評価試験を行った。評価試験1と同様に、後述の条件11から15の各々について、31指を光学式タッチセンサによって2000dpiの横480ピクセル縦800ピクセルの画像を1指あたり5~10枚取得し、1枚を登録画像、他を照合画像としてROCを算出し、認証精度の比較を行った。条件11は、図16のS200の処理で抽出されたペア候補を絞り込まない条件(S202からS204の処理を実行しない条件)である。条件12は、S200の処理で抽出されたペア候補を画像情報(回転角)を用いて絞り込む条件(S202の処理を実施し、S203及びS204の処理を実施しない条件)である。条件13は、S200の処理で抽出されたペア候補を画像情報(回転角)及びスコアを用いて絞り込む条件(S202、S203の処理を実施し、S204の処理を実施しない条件)である。条件14は、S200の処理で抽出されたペア候補を複数組のペア候補の位置関係を用いて絞り込む条件(S202及びS203の処理を実施せず、S204の処理を実施する条件)である。条件15は、S200の処理で抽出されたペア候補をS202からS204の処理で絞り込む条件である。条件11から15の結果を各々、図25の結果55から59に示す。図25に示すように、条件11から15を比較すると、条件11(結果55)に比べ条件12(結果56)は、認証性能が優れていた。条件12(結果56)に比べ条件13(結果57)は、認証性能が更に優れていた。このことから、S202を実施し、画像情報でペア候補を絞り込むことにより、画像情報でペア候補を絞り込まない場合に比べ認証性能を向上させることができ、S202に加えS203の処理を実行することで更に認証性能が向上することが確認された。条件14(結果58)も、条件11(結果55)に比べ認証性能が優れていた。このことから、S204の処理でペア候補を絞り込むことにより、S204でペア候補を絞り込まない場合に比べ認証性能を向上させることができることが確認された。更に、条件12から条件14に比べ、条件15の方が、認証性能が優れていた。このことにより、S202、S203の処理に加え、S204の処理を実行することで、両者が単独で実行される場合に比べ、認証性能を向上させることができることが確認された。
Claims (14)
- プロセッサと、記憶機器とを備えたコンピュータに、
画像を取得する画像取得ステップと、
前記画像取得ステップで取得された前記画像から皮膚の隆線上の汗孔を表す基準点を決定し、当該基準点の前記画像上の位置に対応する情報である位置情報を取得する基準点決定ステップと、
前記基準点決定ステップで取得された複数の前記基準点の中の1つを中心基準点として選択した場合の、前記中心基準点からの距離が所定値未満、且つ、前記中心基準点との間にある前記皮膚の隣合う前記隆線の間となる谷の数が1以下である所定個の前記基準点を所定条件に基づき周辺基準点として抽出する周辺基準点抽出ステップと、
前記複数の基準点の各々について、前記中心基準点の前記位置情報と、前記周辺基準点抽出ステップで抽出された前記所定個の前記周辺基準点の各々の前記画像上の配置の特徴を表す属性情報とを関連付けた汗孔関連情報を生成する情報生成ステップと、
前記情報生成ステップで生成され前記複数の基準点の各々に関する前記汗孔関連情報を、皮膚認証に用いられる認証情報として前記記憶機器に記憶させる記憶制御ステップと
を実行させる為の指示を含む認証情報処理プログラム。 - 前記基準点決定ステップは、前記画像中の前記汗孔の重心を前記基準点として決定し、当該基準点の前記位置情報を取得することを特徴とする請求項1に記載の認証情報処理プログラム。
- 前記所定条件は、既に選択された前記周辺基準点と前記中心基準点とを結んだ線分と、前記周辺基準点の候補となる前記基準点と前記中心基準点とを結んだ線分とがなす角の角度が所定角度以上となる前記基準点を前記中心基準点からの距離が近い順に前記所定個選択する条件であることを特徴とする請求項1又は2に記載の認証情報処理プログラム。
- 前記属性情報は、
前記所定個の周辺基準点の内の前記中心基準点と同じ前記隆線上の前記周辺基準点の数を表す情報と、
前記中心基準点と、前記所定個の周辺基準点の各々とを結んだ場合の所定個の線分の各々を放射線分とし、隣合う2つの前記放射線分上の前記周辺基準点を前記中心基準点周りに順に結んだ所定個の線分の各々を周回線分とした場合の、前記隆線上にある前記周回線分の数を表す情報と
を含む分類情報を含むことを特徴とする請求項1から3の何れかに記載の認証情報処理プログラム。 - 前記属性情報は、前記中心基準点と、前記周辺基準点とが同じ隆線上にあるかを前記周辺基準点の各々について示す情報である放射情報を含むことを特徴とする請求項1から4の何れかに記載の認証情報処理プログラム。
- 前記属性情報は、前記中心基準点と、前記所定個の周辺基準点の各々とを結んだ場合の所定個の線分の各々を放射線分とし、隣合う2つの前記放射線分上の前記周辺基準点を前記中心基準点周りに順に結んだ所定個の線分の各々を周回線分とした場合の、前記周回線分が同じ前記隆線上にあるかを当該周回線分の起点となる前記周辺基準点の各々について示す情報である周回情報を含むことを特徴とする請求項1から5の何れかに記載の認証情報処理プログラム。
- 前記情報生成ステップは、前記複数の基準点の各々について、前記中心基準点の前記位置情報と、前記属性情報とに加え、前記所定個の周辺基準点の前記位置情報に基づく情報である周辺情報を含む前記汗孔関連情報を生成することを特徴とする請求項1から6の何れかに記載の認証情報処理プログラム。
- 前記情報生成ステップは、前記周辺情報として、前記所定個の周辺基準点の各々についての、前記中心基準点との距離を含む前記汗孔関連情報を生成することを特徴とする請求項7に記載の認証情報処理プログラム。
- 前記情報生成ステップは、前記周辺情報として、前記所定個の周辺基準点の各々についての、前記中心基準点と前記周辺基準点とを結ぶ線分の所定方向に対する角度を含む前記汗孔関連情報を生成することを特徴とする請求項7又は8に記載の認証情報処理プログラム。
- 前記情報生成ステップは、前記周辺情報の配列順序と、前記属性情報の配列順序を一致させ、前記所定個の周辺基準点の内、前記中心基準点と同じ隆線上にあり、且つ、前記中心基準点との間の距離が最も長い離間基準点の前記配列順序を1番とし、前記画像上の配置に基づき前記中心基準点を中心とする前記離間基準点から所定方向周りに2番以降の前記配列順序を設定することを特徴とする請求項7から9の何れかに記載の認証情報処理プログラム。
- 前記情報生成ステップで生成された照合用の前記認証情報と、前記記憶機器に記憶されている登録用の前記認証情報との内、前記基準点に関連付けられた前記属性情報が一致する照合用の前記基準点と登録用の前記基準点とのペアを、前記照合用の認証情報と、前記登録用の認証情報との対応を比較する対象の候補となるペア候補として抽出するペア候補抽出ステップと、
前記ペア候補抽出ステップで抽出された前記ペア候補を用いて、前記照合用の認証情報と前記登録用の認証情報との対応を決定する対応決定ステップと
を実行させる為の指示を更に含むことを特徴とする請求項1から10の何れかに記載の認証情報処理プログラム。 - 前記情報生成ステップで生成された照合用の前記認証情報と、前記記憶機器に記憶されている登録用の前記認証情報との内、前記基準点に関連付けられた前記属性情報が一致する照合用の前記基準点と登録用の前記基準点とのペアを、前記照合用の認証情報と、前記登録用の認証情報との対応を比較する対象の候補となるペア候補として抽出するペア候補抽出ステップと、
前記ペア候補抽出ステップで抽出された前記ペア候補について、前記汗孔関連情報の内の前記周辺情報を比較して前記照合用の認証情報と前記登録用の認証情報との対応を決定する対応決定ステップと
を実行させる為の指示を更に含むことを特徴とする請求項7から9の何れかに記載の認証情報処理プログラム。 - 前記対応決定ステップで決定された前記照合用の認証情報と、前記登録用の認証情報との対応を用いて前記照合用の認証情報と、前記登録用の認証情報との類似度を算出する類似度算出ステップを実行させる為の指示を更に含むことを特徴とする請求項11又は12に記載の認証情報処理プログラム。
- プロセッサと、
記憶機器とを備え、
前記プロセッサは、
画像を取得する画像取得手段と、
前記画像取得手段により取得された前記画像から皮膚の隆線上の汗孔を表す基準点を決定し、当該基準点の前記画像上の位置に対応する情報である位置情報を取得する基準点決定手段と、
前記基準点決定手段により取得された複数の前記基準点の中の1つを中心基準点として選択した場合の、前記中心基準点からの距離が所定値未満、且つ、前記中心基準点との間にある前記皮膚の隣合う前記隆線の間となる谷の数が1以下である所定個の前記基準点を所定条件に基づき周辺基準点として抽出する抽出手段と、
前記複数の基準点の各々について、前記中心基準点の前記位置情報と、前記抽出手段により抽出された前記所定個の前記周辺基準点の各々の前記画像上の配置の特徴を表す属性情報とを関連付けた汗孔関連情報を生成する情報生成手段と、
前記情報生成手段により生成され前記複数の基準点の各々に関する前記汗孔関連情報を、皮膚認証に用いられる認証情報として前記記憶機器に記憶させる記憶制御手段と
して機能することを特徴とする認証情報処理装置。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201880030080.1A CN110622170B (zh) | 2017-05-09 | 2018-04-18 | 认证信息处理方法及认证信息处理装置 |
JP2019517533A JP6879524B2 (ja) | 2017-05-09 | 2018-04-18 | 認証情報処理プログラム及び認証情報処理装置 |
KR1020197034071A KR102288696B1 (ko) | 2017-05-09 | 2018-04-18 | 인증 정보 처리 프로그램 및 인증 정보 처리 장치 |
US16/678,348 US11106890B2 (en) | 2017-05-09 | 2019-11-08 | Authentication information processing method, authentication information processing device, and non-transitory computer-readable medium |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2017-093166 | 2017-05-09 | ||
JP2017093166 | 2017-05-09 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US16/678,348 Continuation US11106890B2 (en) | 2017-05-09 | 2019-11-08 | Authentication information processing method, authentication information processing device, and non-transitory computer-readable medium |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2018207571A1 true WO2018207571A1 (ja) | 2018-11-15 |
Family
ID=64105180
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2018/015999 WO2018207571A1 (ja) | 2017-05-09 | 2018-04-18 | 認証情報処理プログラム及び認証情報処理装置 |
Country Status (5)
Country | Link |
---|---|
US (1) | US11106890B2 (ja) |
JP (1) | JP6879524B2 (ja) |
KR (1) | KR102288696B1 (ja) |
CN (1) | CN110622170B (ja) |
WO (1) | WO2018207571A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023281563A1 (ja) * | 2021-07-05 | 2023-01-12 | 日本電気株式会社 | 情報処理システム、情報処理方法及び記憶媒体 |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7251620B2 (ja) * | 2019-05-28 | 2023-04-04 | 日本電気株式会社 | 情報処理装置、情報処理方法及びプログラム |
KR102487919B1 (ko) * | 2021-01-21 | 2023-01-12 | 주식회사 뮤즈라이브 | 보안성을 향상시킨 비가청 음파 통신 기반의 인증 정보 전송 장치 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08509562A (ja) * | 1993-04-27 | 1996-10-08 | パーソナル バイオメトリック エンコーダーズ リミティド | 指紋センサ |
JP2004530217A (ja) * | 2001-05-30 | 2004-09-30 | ギーゼッケ ウント デフリエント ゲーエムベーハー | 指紋検査方法 |
JP2015523876A (ja) * | 2012-05-30 | 2015-08-20 | スコット マクナルティ | 生体測定情報の電磁検出及び分析のためのシステム、方法、及び装置 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4522043B2 (ja) | 2002-09-06 | 2010-08-11 | セイコーエプソン株式会社 | 情報装置及び表示制御方法 |
ATE381739T1 (de) * | 2003-08-29 | 2008-01-15 | Koninkl Philips Electronics Nv | Biometrische identifikationseinrichtung |
JP2013074461A (ja) * | 2011-09-28 | 2013-04-22 | Casio Comput Co Ltd | 画像処理装置、画像処理方法及びプログラム |
JP6526494B2 (ja) | 2015-06-25 | 2019-06-05 | 株式会社ディー・ディー・エス | 情報処理プログラム及び情報処理装置 |
-
2018
- 2018-04-18 KR KR1020197034071A patent/KR102288696B1/ko active IP Right Grant
- 2018-04-18 JP JP2019517533A patent/JP6879524B2/ja active Active
- 2018-04-18 CN CN201880030080.1A patent/CN110622170B/zh active Active
- 2018-04-18 WO PCT/JP2018/015999 patent/WO2018207571A1/ja active Application Filing
-
2019
- 2019-11-08 US US16/678,348 patent/US11106890B2/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08509562A (ja) * | 1993-04-27 | 1996-10-08 | パーソナル バイオメトリック エンコーダーズ リミティド | 指紋センサ |
JP2004530217A (ja) * | 2001-05-30 | 2004-09-30 | ギーゼッケ ウント デフリエント ゲーエムベーハー | 指紋検査方法 |
JP2015523876A (ja) * | 2012-05-30 | 2015-08-20 | スコット マクナルティ | 生体測定情報の電磁検出及び分析のためのシステム、方法、及び装置 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023281563A1 (ja) * | 2021-07-05 | 2023-01-12 | 日本電気株式会社 | 情報処理システム、情報処理方法及び記憶媒体 |
Also Published As
Publication number | Publication date |
---|---|
KR20190139986A (ko) | 2019-12-18 |
US11106890B2 (en) | 2021-08-31 |
JPWO2018207571A1 (ja) | 2020-03-12 |
CN110622170B (zh) | 2023-05-26 |
CN110622170A (zh) | 2019-12-27 |
KR102288696B1 (ko) | 2021-08-11 |
JP6879524B2 (ja) | 2021-06-02 |
US20200074144A1 (en) | 2020-03-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110852160B (zh) | 以图像为基准的生物识别系统及计算机实施方法 | |
US10740589B2 (en) | Skin information processing method, skin information processing device, and non-transitory computer-readable medium | |
EP1374144A1 (en) | Non-contact type human iris recognition method by correction of rotated iris image | |
US11315341B2 (en) | Information processing apparatus, verification method, and computer-readable recording medium recording verification program | |
WO2018207571A1 (ja) | 認証情報処理プログラム及び認証情報処理装置 | |
US11586718B2 (en) | Authentication information processing method, authentication information processing device, and non-transitory computer-readable medium | |
US10740590B2 (en) | Skin information processing method, skin information processing device, and non-transitory computer-readable medium | |
KR20160123209A (ko) | 지문 인식 방법 및 장치 | |
US10430629B2 (en) | Non-transitory computer-readable medium storing information processing program and information processing device | |
JP6526494B2 (ja) | 情報処理プログラム及び情報処理装置 | |
WO2018207572A1 (ja) | 認証情報処理プログラム及び認証情報処理装置 | |
JP2007179267A (ja) | パターン照合装置 | |
JP6723546B2 (ja) | 情報処理プログラム及び情報処理装置 | |
JP6683033B2 (ja) | 生体認証装置、生体認証方法、及び生体認証プログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18799316 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2019517533 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 20197034071 Country of ref document: KR Kind code of ref document: A |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 18799316 Country of ref document: EP Kind code of ref document: A1 |