WO2009008550A1 - 登録装置、照合装置、データ構造及び記憶媒体 - Google Patents
登録装置、照合装置、データ構造及び記憶媒体 Download PDFInfo
- Publication number
- WO2009008550A1 WO2009008550A1 PCT/JP2008/062875 JP2008062875W WO2009008550A1 WO 2009008550 A1 WO2009008550 A1 WO 2009008550A1 JP 2008062875 W JP2008062875 W JP 2008062875W WO 2009008550 A1 WO2009008550 A1 WO 2009008550A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- vein
- unit
- parameter
- authentication
- image
- Prior art date
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/02007—Evaluating blood vessel condition, e.g. elasticity, compliance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
- G06V10/435—Computation of moments
-
- 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
- G06V40/14—Vascular patterns
- G06V40/145—Sensors therefor
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/1455—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters
-
- 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
- G06V40/14—Vascular patterns
Definitions
- Registration device verification device, data structure and storage medium
- the present invention relates to a registration device, a verification device, a data structure, and a storage medium, and is suitable for application to, for example, biometric authentication.
- biometric authentication devices can be installed in mobile communication devices. That is important.
- a biometric authentication device has a vein authentication device that uses a finger vein as a target, and this vein authentication device is a pattern that represents the characteristics of veins in a vein image obtained as a result of vein imaging inside the finger. Information is generated, and this is registered in the storage means or collated with the pattern information registered in the storage means.
- Patent Literature 1 Japanese Patent Publication 2 0 0 5-3 1 2 7 4 8
- the waveform state of the luminance histogram acquired at this time is Since this is a completely different state, there is a problem that authentication accuracy is reduced. Disclosure of the invention
- the present invention has been made in view of the above points, and intends to propose a registration device, a verification device, a data structure, and a storage medium that can improve authentication accuracy.
- the present invention provides a registration apparatus, an image acquisition unit that acquires a vein image of a living body vein, and a parameter that is resistant to affinity conversion from a part of the vein image.
- An extraction unit for extracting the parameters and a registration unit for registering the parameters extracted by the extraction unit in the storage means are provided.
- the present invention is also an authentication device, an extraction unit for extracting parameters based on resistance to affine transformation from a part of a vein image of a biological vein input as an authentication target, and a storage unit.
- a reading unit that reads the registered information, a parameter unit, and a determination unit that determines whether the input user of the parameter item is a registrant according to the degree of collation with the registered information. I made it.
- the present invention provides a data structure for use in the process of determining whether or not a computer is a registrant, and is an affection extracted from a portion of a vein image for a registrant's vein.
- the parameter is compared with the parameter extracted from the corresponding part of the vein image input as the authentication target.
- the present invention is a storage medium for storing data used for the process of determining whether or not a computer is a registrant. Including parameters that are resistant to affine transformations extracted from a portion of the vein image for the vein. It is to be compared with overnight.
- the extraction target of the parameter that is resistant to the affine transformation is made a part of the vein image, so that the movement of the vein in the vein image can be achieved without increasing the imaging range.
- a registration device, a verification device, a data structure, and a storage medium that can secure an allowable amount and thus improve the authentication accuracy can be realized.
- FIG. 1 is a block diagram showing the configuration of the authentication device according to the present embodiment.
- FIG. 2 is a block diagram showing the functional configuration of the control unit in the blood vessel registration mode.
- FIG. 3 is a block diagram showing a functional configuration of the control unit in the authentication mode.
- FIG. 4 is a block diagram showing the configuration of the image processing unit.
- FIG. 5 is a schematic diagram showing images before and after the sharpening process.
- FIG. 6 is a schematic diagram showing a sample image.
- FIG. 7 is a schematic diagram showing a state of rotational movement and parallel movement.
- Fig. 8 is a schematic diagram showing the relationship between the moment invariant and the angle.
- Fig. 9 is a schematic diagram showing the extraction range of the moment invariant.
- FIG. 10 is a schematic diagram showing the relationship between movement and moment invariant.
- Fig. 11 is a schematic diagram used to explain the size of the imaging range centering on the branch point.
- Figure 12 is a schematic diagram used to explain the bifurcation points that should be used as the basis for extracting the moment invariant.
- Figure 13 is a schematic diagram showing the data structure.
- FIG. 14 is a block diagram showing the configuration of the determination unit.
- Fig. 15 is a schematic diagram showing an example of the moment invariant extraction range and an image of the branch point that is the basis of the range.
- Figure 16 is a schematic diagram showing the comparison of moment invariants.
- Fig. 17 is a schematic diagram for explaining the geometric correspondence.
- Figure 18 is a flowchart showing the authentication processing procedure.
- FIG. 1 shows the overall configuration of authentication apparatus 1 according to the present embodiment.
- the authentication device 1 is configured by connecting an operation unit 1 1, an imaging unit 1 2, a memory 1 3, an interface 1 4, and a notification unit 1 5 via a bus 1 6 to a control unit 10. Is done.
- the control unit 10 includes a CPU (Central Processing Unit) that controls the entire authentication device 1, an R 0 M (Read Only Memory) that stores various programs and setting information, and a work memory of the CPU. It is configured as a computer including all RAM (Random Access Memory).
- CPU Central Processing Unit
- R 0 M Read Only Memory
- execution command C OM 1 of the user to be registered (hereinafter referred to as a registrant) or a registrant himself or herself is registered.
- Execution command COM2 of a mode for determining the presence or absence of this (hereinafter referred to as the authentication mode) is input from the operation unit 11 in response to a user operation.
- the control unit 10 determines a mode to be executed based on the execution instructions COM1 and COM2, and based on a program corresponding to the determination result, the imaging unit 12, the memory 13 and the interface 1 and 4 and the notification unit 1 5 are appropriately controlled, The vein registration mode or the authentication mode is executed.
- the imaging unit 12 is included in a wavelength range (70 0 [nm] to 900 [nm]) that has specific absorption characteristics for both deoxygenated hemoglobin and oxygenated hemoglobin.
- Light of a wavelength (hereinafter referred to as near-infrared light) is irradiated onto a surface (hereinafter referred to as a finger placement surface) on which a finger is to be placed.
- the imaging unit 12 uses an image (hereinafter referred to as a vein image) related to an image projected from a vein in a living body part arranged on a finger placement surface (hereinafter referred to as a vein image data). And the vein image data is output to the control unit 10.
- a vein image an image related to an image projected from a vein in a living body part arranged on a finger placement surface
- the memory 13 is a flash memory, for example, and stores or reads data specified by the control unit 10.
- the in-event interface 14 is adapted to exchange various data with an external device connected via a predetermined transmission line.
- the notification unit 15 includes a display unit 15 a and a voice output unit 15 b, and the display unit 15 5 a displays characters and figures based on display data provided from the control unit 10 on the display screen.
- the audio output unit 15 b is configured to output audio based on the audio data given from the control unit 10 from the speaker.
- control unit 10 determines the vein registration mode as the mode to be executed, the control unit 10 notifies that the finger should be placed on the finger placement surface via the notification unit 15, and as shown in FIG. In addition, it functions as a drive unit 21, an image processing unit 2 2, and a registration unit 2 3.
- the drive unit 21 acquires the vein image data by driving the imaging unit 12. That is, the drive unit 21 irradiates near-infrared light on the finger placement surface by driving the light source in the imaging unit 12.
- the driving unit 21 adjusts the lens position of the optical lens in the imaging unit 12 so that the subject is in focus. Furthermore, the drive unit 21 has a predetermined exposure value (EV (Exposure Based on (Value)), the aperture value of the aperture in the imaging unit 12 is adjusted and the shirt evening speed (exposure time) for the imaging device is adjusted.
- EV Exposure Based on (Value)
- the image processing unit 22 extracts from the vein image data given by the image pickup unit 12 as a result of the image pickup by the image pickup unit 12 2, a parameter that is resistant to affine transformation as a feature of the vein image. It is made to do.
- a parameter that is resistant to the affine transformation is a parameter that does not change even if the position changes if the luminance state in the image does not change.
- this parameter will be referred to as a feature value.
- the registration unit 23 generates the data for identifying the registrant based on the feature amount extracted by the image processing unit 22 (hereinafter referred to as “registration data”). Is registered by storing in memory 13.
- control unit 10 can execute the vein registration mode.
- control unit 10 determines the authentication mode as the mode to be executed, the control unit 10 notifies that the finger should be placed on the finger placement surface via the notification unit 15, and then assigns the same symbol to the corresponding part to FIG. 2. As shown in FIG. 3, it functions as a drive unit 21, an image processing unit 2 2, a reading unit 31, and an authentication unit 3 2.
- the drive unit 21 drives the imaging unit 12, and the image processing unit 22 extracts the feature amount based on the vein image data provided from the imaging unit 12.
- the reading unit 31 is obtained by reading the registration data stored in the memory 13 and gives it to the authentication unit 32.
- the authentication unit 3 2 collates the feature amount of the registered data given to the reading unit 31 with the feature amount extracted by the image processing unit 22, and according to the degree of the collation, the vein image Determine whether the data provider can approve of the registrant.
- the certification unit 3 2 determines that it cannot be approved as a registrant, The fact that the approval cannot be made is notified visually and audibly via the display unit 15 a and the audio output unit 15 b.
- the authentication unit 3 2 determines that it can be approved as a registrant, the authentication unit 3 2 sends a device for starting execution of a predetermined process to the device connected to the interface 14. To send.
- a predetermined process for example, a process that should be executed when authentication is successful, such as closing a door for a certain period of time or canceling the operation mode to be restricted, is performed.
- control unit 10 can execute the authentication mode.
- the image processing unit 22 includes a sharpening unit 41, a reference point detection unit 42, and a feature amount extraction unit 43.
- the sharpening unit 41 performs a sharpening process called a LoG filter on the vein image data obtained from the imaging unit 12 (Fig. 2), thereby removing veins included in the vein image. Make it float.
- Figure 5 shows the images before and after the sharpening process.
- the sharpening process clearly distinguishes the difference between the veins and the background (boundary part) in the vein image, and as a result, the veins included in the vein image are highlighted from the background.
- the reference point detection unit 42 detects a branch point of the vein included in the vein image based on the vein image data given from the sharpening unit 41.
- the reference point detection unit 4 2 binarizes the vein image and extracts the center of the vein width or the luminance peak in the binarized image as a previous step for detecting the branch point.
- the line width of the vein is 1 pixel. Generate a pattern. Therefore, the reference point detection unit 42 can uniformly detect the position of the intersection of the veins as compared with the case where the previous process is not performed.
- the feature amount extraction unit 43 acquires the vein image data after the sharpening process from the sharpening unit 41 and calculates the feature amount of the vein image. This feature quantity extraction unit 43 uses a moment invariant as a feature quantity.
- Equation (1) “p” represents the weight in the X-axis direction, and “q” represents the weight in the Y-axis direction. Therefore, as the value of “p” increases, the weight for dispersion in the X-axis direction increases, and as the value of “q” increases, the weight for dispersion in the Y-axis direction increases.
- “w” in equation (1) means the number of pixels in the X-axis direction, and “h” means the number of pixels in the Y-axis direction.
- 1 x l Can be defined by This is a moment considering the distance from the center of gravity, not the distance from the origin. In other words, this value represents the variance considering the center of gravity of the pixel values in the image with respect to the X-axis and Y-axis directions.
- the first-order moment invariant I i is the sum of the variance in the X-axis direction and the variance in the Y-axis direction, as is clear from Eq. (4).
- this moment invariant was proposed by Hu. Vl_K. Hu. Visual pattern recognition by moment invariants. IRE Trans, on Information Theory, lT-8: pp. )
- the feature quantity extraction unit 43 is limited to extracting a moment invariant from the whole vein image, not a part of the vein image.
- the feature quantity extraction unit 43 moves the imaging target regardless of the size of the imaging range (translation of the finger in the plane direction on the finger placement surface). Or the rotation of the finger about the longitudinal axis of the finger), and as a result, the object to be imaged (in comparison with the case where the moment invariant extraction target is the entire vein image)
- the characteristics of veins can be captured appropriately.
- a range in the vicinity of the bifurcation point is adopted as a moment invariant extraction target.
- the feature amount extraction unit 4 3 also acquires the position data of the vein branch point in the image data from the reference inspection extraction unit 4 2.
- the moment invariant is calculated by setting a circular extraction range centered at the bifurcation point indicated in the position data.
- FIG. 10 the first imaging (Fig. 10 (A)) and The left and right vein images of the second image (Fig. 10 (B) and Fig. 10 (C)) moved in the direction of the finger placement surface (moved to the right in the figure) with respect to the image position
- the vein images before and after the sharpening process are shown on the right side, and the enlarged images in the moment invariant extraction range are shown on the lower side.
- the feature quantity extraction unit 43 3 accurately calculates the moment invariant of the branching portion, which has a large individual difference in the vein and is an important element for identification, regardless of the movement in the vein image. Can do.
- the feature quantity extraction unit 43 when setting the extraction range around the branch point, is set so as to include the outline of the branch part.
- the branch point DP detected by the reference point detection unit 42 is an intersection when the vein is 1 pixel wide, but the branch part to be extracted is arbitrary. This is due to the width of
- each branch point on the vein image is at the position from the center and its position.
- the extraction range AR that includes the branch point PX is not set (Fig. 11 (B))
- the individuality becomes poor. Because it loses its significance.
- the radius of the extraction range is equal to or larger than the range (kernel size) used as the sharpening unit in the previous sharpening unit 42.
- the probability of including a branch point PX on the contour line is Because it is expensive.
- the feature quantity extraction unit 4 3 uses only the branch point included in the central region from the center of the vein image to a certain distance as the moment invariant extraction target. Adopted. Therefore, the feature quantity extraction unit 4 3 allows the movement of the imaging target much more than when calculating the moment invariant within the extraction range AR centering on all branch points included in the vein image. It is made to be able to do.
- This registration unit 23 obtains the moment invariant from the feature quantity extraction unit 43, and the number of the moment invariant calculated by the feature quantity extraction unit 43 is equal to or greater than the specified number. Determine if there is.
- the registration unit 23 is considered to be insufficient as the registrant's feature quantity, and the re-imaging is notified via the notification unit 15. A notification has been made.
- the registration unit 2 3 obtains the position of the branch point of the vein in the vein image data from the reference point detection unit 4 2, and Register data including the position of the branch point and the moment ⁇ invariant calculated based on the position is generated.
- Figure 13 shows the data structure of this registration data.
- This identification data has a structure including a header area HAr and a data area DAr.
- this header area H A r the number of branch points and the number of applied orders among the seven orders in the moment of the u u invariant are stored. Since these specify the processing contents of the image processing section 22, the contents stored in the data area DA r are meanings as identifiers for proving that the processing results of the image processing section 22 2 are obtained. It has.
- the amount used as the extraction criterion for the moment invariant Stored in a state in which the position of the bifurcation point (xy coordinates) and h (where h is 2, 3, ..., or 7) and the following moment invariants H are associated.
- the registration unit 23 stores the registration data including the processing result in the image processing unit 22 and the content that proves that the processing result has passed through the image processing unit 22 in the memory 1. It is made to register to 3.
- the authentication unit 32 includes a parameter verification unit 51 and a geometric relationship verification unit 52.
- the parameter verification unit 5 1 is a moment invariant (hereinafter referred to as this) stored in the data area DA r (Fig. 1 3) of the registered data read from the memory 1 3 by the reading unit 31. Is referred to as the registration invariant) and the moment invariant extracted by the image processor 2 2 (Fig. 3) (hereinafter referred to as the authentication invariant).
- the parameter overnight matching unit 51 recognizes the correspondence between the registration invariant and the authentication invariant based on, for example, the branch point that is the extraction criterion of the moment invariant.
- the parameter matching unit 51 also obtains each rate of change d as a result of verification. It is determined whether or not all of these change rates d ⁇ ⁇ are less than a predetermined threshold value. Here, if any one of the change rates di is greater than or equal to the threshold value, this means that the vein image at the time of registration is different from the vein image at the time of authentication. Means not. In this case, the parameter verification unit 51 notifies the registrant that it cannot be approved via the notification unit 15.
- the parameter verification unit 51 notifies the geometric relationship verification unit 52 that processing should be started.
- FIG. 16 shows the processing result in the above-mentioned parameter overnight matching unit 51.
- This FIG. 15 shows the vein image acquired in the vein registration mode in FIG. 15 (A) and the enlarged image in the registration invariant calculation part for the vein image, and in FIG. 15 (B).
- the vein image acquired in the authentication mode from the same part of the registrant and the enlarged image in the calculation part of the authentication invariant relating to the vein image are shown. '
- this figure 15 shows the vein part of the vein image acquired in the vein registration mode ( Figure 15).
- Figure 16 shows that the invariant for registration, the invariant for authentication, and the invariant for authentication were obtained from the same part of the same person, so that there is almost no change in these invariants.
- the geometric relationship verification unit 5 2 should be able to start processing. When notified from 1, verify whether the branch point that is the registration reference point for registration invariants and the branch point that is the extraction reference point for authentication invariants are in a correspondence relationship.
- the geometric relation verification unit 52 searches for a combination of registration invariant and authentication invariant that minimizes the sum of squares of the rate of change, and determines whether the number of combinations is equal to or greater than a specified number. Determine whether.
- the geometric relationship verification unit 52 notifies the registrant that it cannot be approved via the notification unit 15.
- the geometric relation verification unit 52 will register the invariant for registration, which is a combination that satisfies the minimum value of the sum of squares of the change rate below a predetermined threshold, and authentication.
- the position of the branch point that is the extraction reference point for the invariant is acquired. That is, the geometric relation verification unit 52 obtains the position of the branch point corresponding to the registration invariant from the data area DA r (Fig. 13), and the position of the branch point corresponding to the authentication invariant. Is obtained from the image processing unit 22.
- the geometric relationship verification unit 52 obtains the distance between each branch point that is one of the combinations and the distance between the branch points that are the other of the combinations, and obtains the difference between these distances for each combination. Further, the geometric relationship verification unit 52 divides the sum of the differences between the distances obtained for each combination by the number of combinations, and determines whether or not the division result has a geometric correspondence. An evaluation value of
- FIG. 17 shows vein registration. Indicates the combination of numbers assigned to branch points in the vein image acquired in the mode (Fig. 15 (A)) (that is, the combination of branch points).
- the fourth column from the leftmost column is the authentication mode.
- the geometric relationship verification unit 52 calculates the evaluation value, it determines whether or not the evaluation value is less than a predetermined threshold value.
- a predetermined threshold value if the evaluation value is equal to or greater than the predetermined threshold, this means that the value of the registration invariant and the authentication invariant did not change, but the branch point that is the basis for extracting these moment invariants.
- the shape of the veins in the vein image extracted from the invariant for registration and the invariant for authentication is different, and the image that was acquired in this authentication mode Means other than the normal vein image. In this case, the geometric relationship verification unit 52 notifies the registrant that it cannot be approved via the notification unit 15.
- the geometric relation verification unit 52 when the evaluation value is less than the predetermined threshold value, this means that the input person who inputs the vein image in the authentication mode is the registrant.
- the geometric relation verification unit 52 generates a data for starting execution of a predetermined process related to the authentication success, and can be used as an interface that can be approved with the registrant. 4 ( Figure 1) is sent to the device connected.
- step SP 2 the authentication processing procedure in the authentication unit 32 will be described with reference to the flowchart of FIG. That is, when the authentication unit 3 2 receives the authentication mode execution command COM 2 from the operation unit 11 1, the authentication unit 3 2 starts this authentication processing procedure RT, and in step SP 1, the data area DA r (FIG. 13) The rate of change between each registration invariant stored in and the corresponding authentication invariant is calculated for each order, Go to step SP 2.
- step SP2 the authentication unit 32 determines whether all the change rates calculated in step SP1 are less than a predetermined threshold, and all the change rates are less than a predetermined threshold. If yes, go to Step SP 3.
- step SP3 the authentication unit 32 searches for a combination of the registration invariant and the authentication invariant that minimizes the sum of squares of the change rate, and the number of the combination is equal to or greater than the specified number. If the number of combinations exceeds the specified number, go to step SP4.
- step SP4 the authentication unit 3 2 determines a branch point that serves as an extraction reference point between the registration invariant and the authentication invariant that are combinations in which the minimum value of the sum of squares of the change rate satisfies the predetermined threshold value.
- step SP5 an evaluation value is obtained based on the recognition result.
- the authentication unit 32 calculates the distance between each branch point in the registration invariant recognized in step SP 4 and the distance between each branch point in the authentication invariant, and then determines each distance. For each combination searched for in step SP3. Then, the authentication unit 32 uses the value obtained by dividing the total sum of the distances obtained for each combination by the number of combinations searched in step SP 3 as an evaluation value.
- the authentication unit 32 obtains an evaluation value at step SP5, it proceeds to the subsequent step SP6 and determines whether the evaluation value falls below a predetermined threshold value.
- the authentication unit 32 evaluates that the authentication is successful. In this case, the authentication unit 32 proceeds to step SP 7 and executes a predetermined process corresponding to the authentication success, and then ends the authentication processing procedure RT.
- Step SP2 if the evaluation value is greater than or equal to a predetermined threshold, or if at least one change rate between each registration invariant and the corresponding authentication invariant is greater than or equal to the predetermined threshold in Step SP2, Or change in step SP 3. If the number of combinations of registration invariants and authentication invariants that minimize the sum of squares of the conversion rates is less than the specified number, the authentication unit 32 evaluates that the authentication has failed. In this case, the authentication unit 32 proceeds to step SP8, and after executing a predetermined process corresponding to the authentication failure, ends the authentication processing procedure RT.
- the authentication unit 32 can determine the presence or absence of a registrant by comparing the values of feature quantities (moment invariant, position information) without going through image matching processing.
- the authentication apparatus 1 extracts feature quantities having parameters that are resistant to affine transformation only in the vein branch portion displayed in the vein image (for example, FIG. 10). As a result, the authentication device 1 can extract the equivalent feature amount without increasing the imaging range in the imaging unit 12 even when the vein moves in the vein image.
- this authentication device 1 it is possible to prevent a decrease in authentication accuracy while securing a vein movement allowable amount in the vein image, as compared with the case where the vein feature amount in the entire vein image is extracted. . For this reason, it is particularly useful when it is installed in a device such as a mobile phone that has a strong demand for downsizing.
- moment ⁇ invariant as the feature quantity, it is possible to translate the finger in the plane direction on the finger placement surface, or to rotate the finger around the longitudinal axis of the finger. Even if there is both movements, it is possible to extract equivalent features.
- each branch portion included in the central area CAR (FIG. 12) from the center position of the vein image to a certain distance is extracted as the center.
- a range is set, and feature quantities are extracted from the set extraction range.
- the feature extraction target is the central area CAR.
- the structure of the imaging unit 12 is such that a light source and an imaging element are arranged so as to sandwich a finger placed on a finger placement surface (for example, Japanese Patent Laid-Open No. 2 0 5-3 5 3 0 1 4, When the light source and the image sensor are arranged on the same direction side with respect to the finger placement surface (for example, the same publication, Fig. 2) Is particularly useful.
- the near infrared light emitted from the light source to the vein from the end of the finger placement surface This is because the edge of the vein image in which the visible light sometimes enters as noise and the vein is difficult to be reflected due to this noise can be excluded from the feature extraction target.
- the feature quantity having parameters that are resistant to the affine transformation is extracted only at the branching portion of the vein displayed in the vein image, so that the imaging range does not have to be increased. Therefore, it is possible to secure the allowable amount of vein movement in the vein image, and thus to realize the authentication device 1 capable of improving the authentication accuracy.
- the log file processing is applied to the sharpening process in the sharpening unit 41 .
- the present invention is not limited to this.
- the morphology Various other filter events such as filters, Laplacian Phil evening, Gaussian Phil evening or Sobel Phil evening can be applied.
- the number of fills to be applied may be 2 or more.
- the branch point is applied as the detection target in the reference point detection unit 42 .
- the present invention is not limited to this, and the end point, the bending point, or these A combination of the above may be applied.
- the points on the vein other than all of the end points, branch points, and inflection points, such as the points that intersect the circle centered on the vein image among the points that make up the vein You may make it apply.
- the reference point detection unit 42 has described the case where the detection target is fixed to the branch point and detected, the present invention is not limited to this, and the detection target is switched at a predetermined timing. May be.
- the detection target is switched in the order of a branch point, a bending point, and a point intersecting with a circle centering on the image center position.
- the radius of the circle centered on the center position of the image is switched, and a point intersecting with the circle is detected.
- an identifier for specifying the detection method (that is, the detection method) is described in the header area HA r (Fig. 13), and the detection method represented by the identifier is displayed in the authentication mode.
- this authentication device 1 it is possible to prevent the masquerading as a registrant based on the stolen information, and the authentication accuracy can be further improved.
- the moment invariant of Hu is applied as an extraction target in the feature amount extraction unit 43 .
- the present invention is not limited to this, for example, Zernike moment, entropy, etc. may be applied.
- it is invariant with respect to both movements of the finger parallel movement in the surface direction on the finger placement surface, or rotational movement of the finger centered on the longitudinal axis of the finger.
- a variable that is invariant to translation only such as a local luminance histogram in a vein image.
- various other features can be applied if they are resistant to affine transformation. wear.
- the authentication device 1 having the imaging function, the verification function, and the registration function has been described.
- the present invention is not limited to this, and the function is determined according to the application. Alternatively, a part of each function may be applied to a single device.
- the present invention can be used in the field of biometric authentication.
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- General Physics & Mathematics (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Multimedia (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Vascular Medicine (AREA)
- Human Computer Interaction (AREA)
- Physiology (AREA)
- Cardiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Collating Specific Patterns (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP08778226A EP2169621A1 (en) | 2007-07-11 | 2008-07-10 | Registering apparatus, collaiting apparatus, data structure, and storage medium |
CN200880023234A CN101730905A (zh) | 2007-07-11 | 2008-07-10 | 登记装置、核对装置、数据结构和存储介质 |
US12/667,295 US20110007943A1 (en) | 2007-07-11 | 2008-07-10 | Registration Apparatus, Checking Apparatus, Data Structure, and Storage Medium (amended |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2007-182631 | 2007-07-11 | ||
JP2007182631A JP2009020698A (ja) | 2007-07-11 | 2007-07-11 | 登録装置、照合装置、データ構造及び記憶媒体 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2009008550A1 true WO2009008550A1 (ja) | 2009-01-15 |
Family
ID=40228713
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2008/062875 WO2009008550A1 (ja) | 2007-07-11 | 2008-07-10 | 登録装置、照合装置、データ構造及び記憶媒体 |
Country Status (7)
Country | Link |
---|---|
US (1) | US20110007943A1 (ja) |
EP (1) | EP2169621A1 (ja) |
JP (1) | JP2009020698A (ja) |
KR (1) | KR20100038190A (ja) |
CN (1) | CN101730905A (ja) |
TW (1) | TW200912768A (ja) |
WO (1) | WO2009008550A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012131993A (ja) * | 2010-11-30 | 2012-07-12 | Sumitomo Chemical Co Ltd | 高分子化合物及びその製造方法、並びに発光素子 |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5729029B2 (ja) * | 2011-03-10 | 2015-06-03 | セイコーエプソン株式会社 | 識別装置及び識別方法 |
CN102542569B (zh) * | 2011-12-21 | 2015-03-11 | 武汉市兑尔科技有限公司 | 一种快速图像配准及其标定方法及实现其的系统 |
CN108205643B (zh) * | 2016-12-16 | 2020-05-15 | 同方威视技术股份有限公司 | 图像匹配方法和装置 |
CN110119724A (zh) * | 2019-05-16 | 2019-08-13 | 天津科技大学 | 一种指静脉识别方法 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0248776A (ja) * | 1988-08-10 | 1990-02-19 | Fujitsu Ltd | 指紋照合方法および装置 |
JPH06131444A (ja) * | 1992-08-19 | 1994-05-13 | Masanori Sugisaka | モーメント不変量を用いた天然産物またはその加工品の形状識別方法 |
JP2005353014A (ja) | 2004-05-14 | 2005-12-22 | Sony Corp | 撮像装置 |
JP2006053773A (ja) * | 2004-08-12 | 2006-02-23 | Sony Corp | 画像処理方法およびその装置 |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5835641A (en) * | 1992-10-14 | 1998-11-10 | Mitsubishi Denki Kabushiki Kaisha | Image pick-up apparatus for detecting and enlarging registered objects |
US6106464A (en) * | 1999-02-22 | 2000-08-22 | Vanderbilt University | Apparatus and method for bone surface-based registration of physical space with tomographic images and for guiding an instrument relative to anatomical sites in the image |
US20070036400A1 (en) * | 2005-03-28 | 2007-02-15 | Sanyo Electric Co., Ltd. | User authentication using biometric information |
JP5003092B2 (ja) * | 2006-10-17 | 2012-08-15 | ソニー株式会社 | 登録装置、照合装置、認証方法及び認証プログラム |
US8317705B2 (en) * | 2008-12-10 | 2012-11-27 | Tomtec Imaging Systems Gmbh | Method for generating a motion-corrected 3D image of a cyclically moving object |
-
2007
- 2007-07-11 JP JP2007182631A patent/JP2009020698A/ja active Pending
-
2008
- 2008-05-30 TW TW097120335A patent/TW200912768A/zh unknown
- 2008-07-10 WO PCT/JP2008/062875 patent/WO2009008550A1/ja active Application Filing
- 2008-07-10 CN CN200880023234A patent/CN101730905A/zh active Pending
- 2008-07-10 US US12/667,295 patent/US20110007943A1/en not_active Abandoned
- 2008-07-10 EP EP08778226A patent/EP2169621A1/en not_active Withdrawn
- 2008-07-10 KR KR1020107000289A patent/KR20100038190A/ko not_active Application Discontinuation
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0248776A (ja) * | 1988-08-10 | 1990-02-19 | Fujitsu Ltd | 指紋照合方法および装置 |
JPH06131444A (ja) * | 1992-08-19 | 1994-05-13 | Masanori Sugisaka | モーメント不変量を用いた天然産物またはその加工品の形状識別方法 |
JP2005353014A (ja) | 2004-05-14 | 2005-12-22 | Sony Corp | 撮像装置 |
JP2006053773A (ja) * | 2004-08-12 | 2006-02-23 | Sony Corp | 画像処理方法およびその装置 |
Non-Patent Citations (1)
Title |
---|
M-K HU: "Visual pattern recognition by moment invariants", IRE TRANS. ON INFORMATION THEORY, vol. IT-8, 1962, pages 179 - 187 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012131993A (ja) * | 2010-11-30 | 2012-07-12 | Sumitomo Chemical Co Ltd | 高分子化合物及びその製造方法、並びに発光素子 |
Also Published As
Publication number | Publication date |
---|---|
EP2169621A1 (en) | 2010-03-31 |
CN101730905A (zh) | 2010-06-09 |
TW200912768A (en) | 2009-03-16 |
KR20100038190A (ko) | 2010-04-13 |
JP2009020698A (ja) | 2009-01-29 |
US20110007943A1 (en) | 2011-01-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6650946B2 (ja) | モバイル・デバイスを用いてキャプチャしたイメージを使用する指紋ベースのユーザ認証を実行するためのシステムおよび方法 | |
JP5061645B2 (ja) | 情報抽出方法、情報抽出装置、プログラム、登録装置及び照合装置 | |
JP4937607B2 (ja) | 生体認証方法及び生体認証装置 | |
JP4762805B2 (ja) | 生体認証方法及び生体認証システム | |
CN106845414B (zh) | 用于生物特征验证的质量度量的方法和系统 | |
JP4547629B2 (ja) | 登録装置、登録方法及び登録プログラム | |
WO2007100145A1 (ja) | 登録装置、照合装置、抽出方法及びプログラム | |
JP6790995B2 (ja) | 照合装置、照合方法および照合プログラム | |
EP2610820A2 (en) | Authentication apparatus, authentication program, and method of authentication | |
CN108389053B (zh) | 支付方法、装置、电子设备及可读存储介质 | |
WO2009008550A1 (ja) | 登録装置、照合装置、データ構造及び記憶媒体 | |
WO2008047935A1 (fr) | Procédé d'identification, dispositif d'enregistrement, dispositif et programme de collation de modèles | |
JP2017211938A (ja) | 生体情報処理装置、生体情報処理方法および生体情報処理プログラム | |
US20220044014A1 (en) | Iris authentication device, iris authentication method and recording medium | |
JP6908843B2 (ja) | 画像処理装置、画像処理方法、及び画像処理プログラム | |
JP5050642B2 (ja) | 登録装置、照合装置、プログラム及びデータ構造 | |
JP5003092B2 (ja) | 登録装置、照合装置、認証方法及び認証プログラム | |
JP6187262B2 (ja) | 生体情報処理装置、生体情報処理方法及び生体情報処理用コンピュータプログラム | |
CN112801067B (zh) | 一种检测虹膜光斑的方法及计算设备 | |
KR20160123209A (ko) | 지문 인식 방법 및 장치 | |
JP7315884B2 (ja) | 認証方法、認証プログラム、および情報処理装置 | |
CN115204893A (zh) | 用于电子支付的人脸识别方法、装置、计算机设备 | |
JP2007179267A (ja) | パターン照合装置 | |
JP5914995B2 (ja) | 生体識別装置、及び、生体識別方法 | |
JP5729029B2 (ja) | 識別装置及び識別方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200880023234.0 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 08778226 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12667295 Country of ref document: US |
|
ENP | Entry into the national phase |
Ref document number: 20107000289 Country of ref document: KR Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2008778226 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |