WO2010146675A1 - 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム - Google Patents
生体認証装置、生体認証方法及び生体認証用コンピュータプログラム Download PDFInfo
- Publication number
- WO2010146675A1 WO2010146675A1 PCT/JP2009/061037 JP2009061037W WO2010146675A1 WO 2010146675 A1 WO2010146675 A1 WO 2010146675A1 JP 2009061037 W JP2009061037 W JP 2009061037W WO 2010146675 A1 WO2010146675 A1 WO 2010146675A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- biometric
- biometric information
- input
- registered
- image
- Prior art date
Links
Images
Classifications
-
- 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
- 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
- G06V40/13—Sensors therefor
- G06V40/1312—Sensors therefor direct reading, e.g. contactless acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/98—Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
- G06V10/993—Evaluation of the quality of the acquired pattern
-
- 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
- G06V40/107—Static hand or arm
-
- 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
-
- 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
- G06V40/50—Maintenance of biometric data or enrolment thereof
- G06V40/55—Performing matching on a personal external card, e.g. to avoid submitting reference information
Definitions
- Embodiments disclosed herein include a biometric authentication device, a biometric authentication method, and a biometric that determine whether or not to authenticate a target person by comparing biometric information represented in a biometric image with biometric information registered in advance.
- the present invention relates to an authentication computer program.
- biometric authentication technology for authenticating a user of an apparatus or system based on a biometric image representing biometric information such as a hand or finger vein pattern, a fingerprint, or a palm print has been developed.
- a biometric authentication device using such a biometric authentication technology acquires, for example, a biometric image representing biometric information of a user who intends to use the biometric authentication device as an input biometric image. Then, the biometric authentication device collates the user's input biometric information represented in the input biometric image with registered biometric information that is biometric information represented in the registered user's biometric image.
- the biometric authentication device determines that the input biometric information and the registered biometric information match based on the result of the verification process, the biometric authentication device authenticates the user as a registered user having a legitimate authority.
- the biometric authentication device permits an authenticated user to use a device in which the biometric authentication device is incorporated or another device connected to the biometric authentication device.
- biometric authentication apparatus acquires an image in which a vein pattern of a user's hand is captured as a biometric image representing biometric information
- the obtained image is obtained if the entire hand is not included in the imaging range of the sensor. Only part of the vein pattern of the hand is visible.
- the image of the hand vein pattern shown in the resulting image will be small, so that the details of the vein pattern on the image will be It will be unclear.
- the biometric authentication device cannot use the information of the missing or unclear portion for the collation. . Therefore, the biometric authentication device can accurately check the degree of coincidence between the vein pattern represented in the previously registered image and the vein pattern represented in the input image with respect to the missing or unclear portion. Disappear. As a result, the matching accuracy is lowered. Therefore, a technique has been developed that determines whether or not a biological image is appropriate and, when it is determined that the biological image is not appropriate, causes the user to re-input the biological image.
- the biometric authentication device On the other hand, if the time required for executing the biometric authentication process becomes long, the biometric authentication device impairs the convenience of the user. If the re-input of the biometric image is required, the time required for executing the biometric authentication process becomes longer.
- the biometric authentication device Must perform multiple biometric authentication processes. Therefore, a biometric authentication device that employs the above-described technology reads the user's biometric information again, and executes a plurality of biometric authentication processes after determining whether the input biometric image generated by reading again is appropriate. become.
- the biometric authentication apparatus takes a long time to execute the biometric authentication process. Then, the greater the number of registered biometric images registered in the biometric authentication device, the greater the number of executions of the biometric authentication process. Therefore, when a large number of registered biometric images are registered in the biometric authentication device, the waiting time of the user until all the biometric authentication processes are completed increases significantly, and as a result, the convenience for the user is significantly impaired. There was a risk of being lost.
- an object of the present specification is to provide a biometric authentication device, a biometric authentication method, and a biometric authentication computer program that can reduce the time required for biometric authentication processing even when a plurality of registered biometric images are registered. .
- a biometric authentication device acquires biometric information of a user, generates a first input biometric image representing the biometric information, and registered biometric information of at least one registered user registered in advance. And a processing unit.
- the processing unit has a pass / fail determination function for determining whether or not the first input biometric image is appropriate in order to use the biometric information represented in the first input biometric image for matching with the registered biometric information; When it is determined that one input biometric image is inappropriate, a selection function for selecting registered biometric information similar to the biometric information represented in the first input biometric image, and the biometric information acquisition unit is a user's biometric A verification processing function for verifying the biometric information represented in the second input biometric image generated by re-acquiring information and the selected registered biometric information is realized.
- a biometric authentication method is provided.
- the user's biometric information is acquired, a first input biometric image representing the biometric information is generated, and the biometric information represented in the first input biometric image is registered in advance. If the first input biometric image is determined to be appropriate for use in the verification with the registered biometric information of the registered user, the registration is performed. Of the biometric information, registered biometric information similar to the biometric information represented in the first input biometric image is selected, and the second input biometric image is generated by re-acquiring the biometric information of the user, and the second Collating the biometric information represented in the input biometric image with the selected registered biometric information.
- a computer performs biometric authentication to determine whether or not to authenticate a user by comparing biometric information of the user with registered biometric information of at least one registered user registered in advance.
- a computer program to be executed. This computer program is used to collate the biometric information of the user represented in the first input biometric image generated by the biometric information acquisition unit with the registered biometric information of at least one registered user stored in the storage unit. In order to do so, it is determined whether or not the first input biometric image is appropriate. If it is determined that the first input biometric image is inappropriate, the first input biometric image is displayed in the registered biometric information. The registered biometric information similar to the biometric information thus selected is selected, and the computer is caused to collate the biometric information represented in the second input biometric image generated by the biometric information acquisition unit with the selected registered biometric information. Has instructions.
- the biometric authentication device, the biometric authentication method, and the biometric authentication computer program disclosed in this specification can reduce the time required for biometric authentication processing even when a plurality of registered biometric images are registered.
- FIG. 1 is a schematic configuration diagram of a biometric authentication device according to one embodiment.
- 2A to 2C are diagrams showing an example of the positional relationship between the imaging region and the hand of the biological information acquisition unit.
- FIGS. 3A to 3C are diagrams showing another example of the positional relationship between the imaging region of the biological information acquisition unit and the hand.
- FIG. 4 is a functional block diagram of a processing unit included in the biometric authentication device, showing functions realized for executing biometric authentication processing for a user according to one embodiment.
- FIG. 5 is a diagram showing an operation flowchart of biometric authentication processing controlled by a computer program executed on the processing unit.
- FIG. 6 is a conceptual diagram showing a comparison between the time required to execute the biometric authentication process according to one embodiment and the time required to execute the biometric authentication process according to the conventional technique.
- this biometric authentication device When performing biometric authentication processing for a user, this biometric authentication device acquires an input biometric image that is an image representing the biometric information of the user. And this biometrics apparatus is the input biometric information which is a user's biometric information represented by the input biometric image with each of the registered biometric information which is biometric information of a plurality of registered users registered in advance in the biometric authentication apparatus. Match. When it is determined that the input biometric information matches any registered biometric information based on the result of the collation process, the biometric authentication device corresponds to the registered biometric information determined to match the input biometric information. Authenticate as a registered user.
- the input biometric information is not clear, for example, the input biometric information represented in the input biometric image is unclear or only a part of the biometric information to be used for matching is shown in the input biometric image. If it is incomplete, the user is instructed to re-enter biometric information. On the other hand, this biometric authentication device selects a registered biometric image in which registered biometric information similar to the incomplete input biometric information is represented from all registered biometric images registered in advance. The biometric authentication device only receives the input biometric information represented in the appropriate input biometric image obtained by causing the user to read the biometric information again, and the registered biometric information represented in the selected registered biometric image. Match. As a result, the biometric authentication apparatus reduces the number of registered biometric images for which biometric authentication processing is executed after re-input of the input biometric image, thereby shortening the processing time required for executing the entire biometric authentication processing.
- the biometric authentication device uses a hand or finger vein pattern as biometric information to be biometrically authenticated.
- the biometric information subject to biometric authentication may be other biometric information such as a fingerprint, palm print, or nose print.
- the term “collation process” is used to indicate a process of calculating a similarity indicating the degree of similarity between input biometric information and registered biometric information.
- biometric authentication process indicates not only the verification process but also the entire authentication process including the process of determining whether to authenticate the user using the similarity obtained by the verification process. Used for.
- FIG. 1 shows a schematic configuration diagram of a biometric authentication apparatus.
- the biometric authentication device 1 includes a display unit 2, a biometric information acquisition unit 3, a storage unit 4, and a processing unit 5.
- the biometric authentication device 1 performs biometric authentication processing using a blood vessel image representing a vein pattern of a user's hand or finger.
- the biometric authentication device 1 authenticates the user as one of registered users registered in advance in the biometric authentication device 1 as a result of the biometric authentication process, the user installs the device on which the biometric authentication device 1 is mounted. Allow to use.
- the biometric authentication device 1 transmits a signal indicating that the user has been authenticated to another device (not shown) and permits the user to use the other device.
- the biometric authentication device 1 may include an input unit 6 such as a keyboard, a mouse, or a touch pad, for example.
- the biometric authentication device 1 acquires the command, data, or user identification information input by the user via the input unit 6, and passes the command, data, or user identification information to the processing unit 5. Also good. However, when the user does not need to input information other than the biometric information to the biometric authentication device 1, the input unit 6 may be omitted.
- the display unit 2 includes a display device such as a liquid crystal display or a CRT monitor. Then, the display unit 2 displays a guidance message for placing a hand or a finger to a user at a position where the biological information acquisition unit 3 can acquire an appropriate blood vessel image. The display unit 2 displays various information related to the application executed by the processing unit 5.
- the biometric information acquisition unit 3 generates an input blood vessel image representing the vein pattern of the user's hand or finger.
- the biological information acquisition unit 3 includes, for example, a two-dimensional sensor and an optical system arranged to form an image of an object arranged in a predetermined imaging region on the two-dimensional sensor. Then, the biometric information acquisition unit 3 passes the generated input blood vessel image to the processing unit 5.
- the biometric information acquisition unit 3 may further include an illumination light source that irradiates the user's hand or finger with near infrared light.
- the display part 2 and the biometric information acquisition part 3 may be formed integrally.
- the storage unit 4 includes, for example, at least one of a semiconductor memory, a magnetic disk device, and an optical disk device. And the memory
- the storage unit 4 stores a program for executing biometric authentication processing. Furthermore, the storage unit 4 stores, for each registered user, data related to a vein pattern of a hand or a specific finger, which is registered biometric information of the registered user.
- the data relating to the registered biometric information can be, for example, a registered blood vessel image that is an image of a vein pattern of either the left or right hand of a registered user or a specific finger.
- the data related to the registered biometric information may be a registered blood vessel image itself or a feature value for collation processing extracted from the partial region.
- the storage unit 4 may store effective biometric information for selection processing, which will be described later, extracted from the registered blood vessel image, separately from the feature amount for verification processing, as one of data relating to each registered biometric information. Good.
- the storage unit 4 stores a guidance message for placing a hand or a finger at a position where the biometric information acquisition unit 3 can acquire an appropriate blood vessel image.
- the processing unit 5 includes one or a plurality of processors and their peripheral circuits. Then, the processing unit 5 executes a biometric authentication process using the input blood vessel image acquired from the biometric information acquisition unit 3 and representing the vein pattern of the user.
- the biometric authentication device 1 cannot execute an accurate matching process.
- FIG. 2A to 2C are diagrams showing an example of the positional relationship between the imaging region of the biological information acquisition unit 3 and the hand.
- the user moves his / her hand in the horizontal direction from the left toward the imaging region 200 of the biological information acquisition unit 3.
- FIG. 2A only about half of the user's hand 210 is in the shooting area 200. Therefore, the input blood vessel image acquired in the state shown in FIG. 2A shows only about half of the vein pattern of the palm portion.
- the processing unit 5 performs the matching process using the input blood vessel image acquired in the state shown in FIG. 2A, information about half of the palm vein pattern is missing. There is a possibility that the vein pattern of the user cannot be accurately matched with the vein pattern of the registered user.
- FIG. 1 the vein pattern of the user cannot be accurately matched with the vein pattern of the registered user.
- the range of the user's hand 210 included in the imaging region 200 is wider than that in FIG. 2A, but the vein pattern near the thumb is still out of the imaging region 200. Yes. Therefore, when the processing unit 5 performs the matching process using the input blood vessel image acquired in the state shown in FIG. 2B, the input blood vessel image acquired in the state shown in FIG. The accuracy of collation can be expected to be better than using. However, the processing unit 5 may not be able to accurately collate the user's vein pattern with the registered user's vein pattern.
- the imaging region 200 includes almost the entire area of the user's hand 210. Therefore, the entire vein pattern of the palm portion is shown in the input blood vessel image acquired in the state shown in FIG. As a result, the processing unit 5 performs the matching process using the input blood vessel image acquired in the state shown in FIG. 2C, so that the user's vein pattern is accurately matched with the registered user's vein pattern. Can be verified.
- FIGS. 3A to 3C are diagrams showing another example of the positional relationship between the imaging region of the biological information acquisition unit 3 and the hand.
- the user moves his / her hand closer to the biological information acquisition unit 3 along a direction orthogonal to the sensor surface of the biological information acquisition unit 3.
- the area of the user's hand 310 occupying the imaging region 300 is less than half of the area when the user's hand is placed at an appropriate position.
- the noise component superimposed on the input blood vessel image increases. Therefore, in the input blood vessel image acquired in the state shown in FIG. 3A, the fine structure of the vein pattern of the palm part becomes unclear.
- the processing unit 5 when the processing unit 5 performs the matching process using the input blood vessel image acquired in the state illustrated in FIG. 3A, information regarding the fine structure of the vein pattern cannot be used. Therefore, the processing unit 5 may not be able to accurately collate the user's vein pattern with the registered user's vein pattern.
- the area of the user's hand 310 occupying the imaging region 300 is larger than that in FIG. 3A, but the area of the hand 310 is an appropriate position of the user's hand. Smaller than the area of the hand when placed on. Therefore, when the processing unit 5 performs the matching process using the input blood vessel image acquired in the state shown in FIG. 3B, the input blood vessel image acquired in the state shown in FIG.
- the processing unit 5 may not be able to accurately collate the user's vein pattern with the registered user's vein pattern.
- FIG. 3C since the user's hand is placed at an appropriate position, the area of the user's hand 310 that occupies the imaging region 300 is sufficiently large. Therefore, in the input blood vessel image acquired in the state shown in FIG. 3C, the fine structure of the vein pattern of the palm portion can be determined. As a result, the processing unit 5 performs the matching process using the input blood vessel image acquired in the state shown in FIG. 3C, so that the user's vein pattern is accurately matched with the registered user's vein pattern. Can be verified.
- the processing unit 5 uses the use represented in the inappropriate input blood vessel image.
- the biometric information related to the user's hand is collated with the biometric information related to the registered user's hand.
- the processing unit 5 includes data related to the vein pattern of the registered user corresponding to the biometric information corresponding to some extent to the biometric information related to the hand represented in the inappropriate input blood vessel image among the data related to the vein patterns of all registered users. Select only. Then, when an input blood vessel image captured with the user's hand placed at an appropriate position is obtained, the processing unit 5 performs a verification process only on data related to the vein pattern of the selected registered user. Execute.
- FIG. 4 is a functional block diagram of the processing unit 5 showing functions realized for executing the biometric authentication process.
- the processing unit 5 includes a quality determination unit 11, a guidance processing unit 12, an effective biological information extraction unit 13, an effective biological information matching unit 14, a selection unit 15, and a feature amount extraction unit. 16, a verification unit 17, and an authentication determination unit 18.
- Each of these units included in the processing unit 5 is a functional module implemented by a computer program executed on a processor included in the processing unit 5.
- these units included in the processing unit 5 may be mounted on the biometric authentication device 1 as firmware.
- the quality determination unit 11 determines whether or not the input blood vessel image generated by the biometric information acquisition unit 3 is appropriate for use in the matching process. For example, the pass / fail judgment unit 11 binarizes the input blood vessel image by comparing the pixel value of each pixel of the input blood vessel image with a binarization threshold value.
- the binarization threshold can be, for example, an average value of pixel values of the input blood vessel image.
- the pass / fail determination unit 11 determines that the pixel value is 2 If it is higher than the threshold value, the pixel is set as a hand region candidate pixel.
- the pass / fail judgment unit 11 performs a labeling process on the hand region candidate pixels obtained for the entire input blood vessel image, and determines a region including the vein pattern used for matching as a region in which the hand region candidate pixels are connected. It is detected as a hand region that is an image of Note that whether the value of the pixel included in the region where the user's hand is shown is higher or lower than the value of the pixel included in the background region is determined by the configuration of the biological information acquisition unit 3. Therefore, when the value of the pixel included in the area where the user's hand is captured is lower than the value of the pixel included in the background area, the pass / fail judgment unit 11 selects a pixel having a pixel value lower than the binarization threshold. A hand region candidate pixel may be used.
- the pass / fail judgment unit 11 obtains a pass / fail judgment index that is an index for determining whether or not the input blood vessel image is appropriate for use in the matching process. If the pass / fail judgment index indicates that the image of the user's hand or finger in the input blood vessel image is so small that it is difficult to discriminate details of the vein pattern, the input blood vessel The image is determined to be inappropriate. In addition, the pass / fail judgment unit 11 indicates that when any pass / fail judgment index indicates that at least a part of the hand or finger image including the vein pattern used for collation is missing in the input blood vessel image, the input blood vessel The image is determined to be inappropriate.
- the pass / fail determination unit 11 determines whether any of the pass / fail determination indices can identify the vein pattern used for matching from the image of the user's hand or finger shown in the input blood vessel image. It is determined that Hereinafter, an input blood vessel image determined to be inappropriate is referred to as an inappropriate input blood vessel image, while an input blood vessel image determined to be appropriate is referred to as an appropriate input blood vessel image.
- the quality determination unit 11 calculates the area and the center of gravity position of the hand region as an example of a quality determination index.
- the quality determination unit 11 may calculate the number of pixels that are in contact with the image edges of the input blood vessel image among the pixels included in the hand region as another example of the quality determination index.
- the pass / fail determination unit 11 determines that the input blood vessel image is inappropriate when the position of the center of gravity of the hand region is within a predetermined distance from any image edge of the input blood vessel image. This predetermined distance may be, for example, half of the average value of the width of the hand region on the input blood vessel image when a human hand or finger is placed at an appropriate position with respect to the biological information acquisition unit 3. it can.
- the pass / fail judgment unit 11 calculates the number of pixels in the hand region that is in contact with one of the left and right image edges of the input blood vessel image and the number of pixels in the hand region that is in contact with the other image edge. If the difference is greater than or equal to a predetermined threshold, the input blood vessel image may be determined to be inappropriate.
- This predetermined threshold can be set to, for example, 1/3 or 1/4 of the length in the vertical direction of the hand region on the input blood vessel image.
- the pass / fail judgment unit 11 has an average area of the hand region on the input blood vessel image when the human hand or finger is placed at an appropriate position with respect to the biological information acquisition unit 3.
- the input blood vessel image may be determined to be inappropriate if it is smaller than the minimum allowable area that is a value obtained by multiplying a certain reference area by a predetermined coefficient.
- This minimum allowable area corresponds to the minimum value of the area of the hand region that can be identified on the input blood vessel image by the details of the vein pattern used for the matching process, for example, a range of values obtained by multiplying the reference area by 0.5 to 0.8 It can be a value included in.
- the quality determination unit 11 determines that the input blood vessel image is inappropriate, the quality determination unit 11 estimates the cause of the inappropriateness. For example, when the barycentric position of the hand region is within a predetermined distance from the left end of the input blood vessel image, the pass / fail determination unit 11 is caused by the position of the user's hand or finger being the biometric information acquisition unit. Estimated to be too far to the right of 3.
- the number of pixels in the hand region in contact with the left end of the input blood vessel image may be more than a predetermined threshold value than the number of pixels in the hand region in contact with the right end of the input blood vessel image.
- the quality determination unit 11 estimates that the cause of the inappropriateness is that the position of the hand or finger is too close to the right side with respect to the biological information acquisition unit 3.
- the pass / fail judgment unit 11 is caused by the position of the hand or the finger relative to the biometric information acquisition unit 3 Estimate that it is too far to the left.
- the number of pixels in the hand region in contact with the right end of the input blood vessel image may be more than a predetermined threshold value than the number of pixels in the hand region in contact with the left end of the input blood vessel image.
- the quality determination unit 11 estimates that the cause of the inappropriateness is that the position of the hand or finger is too close to the left side with respect to the biological information acquisition unit 3. Furthermore, the center of gravity of the hand region and the number of pixels in the hand region that contact the image edge do not meet the criteria for determining the input blood vessel image as inappropriate, and the input blood vessel image is inappropriate because the area of the hand region is small. May be determined. In this case, the quality determination unit 11 estimates that the cause of the inappropriateness is that the position of the hand or finger is too far from the biological information acquisition unit 3.
- the pass / fail judgment unit 11 passes information indicating the pass / fail judgment result to the processing unit 5. If the input / output image is determined to be inappropriate, the pass / fail determination unit 11 passes to the processing unit 5 cause information indicating the cause of the input blood vessel image and information indicating the hand region in the input blood vessel image. .
- the guidance processing unit 12 reads a guidance message corresponding to the cause information notified from the processing unit 5 from the storage unit 4. For example, when the cause information indicates that the position of the hand or finger is too far to the left with respect to the biological information acquisition unit 3, the guidance processing unit 12 reads a guidance message corresponding to the cause from the storage unit 4. .
- the guidance message can be, for example, a message that prompts the user to remove the cause of the inappropriateness of the input blood vessel image, such as “Please move your hand a little to the right and try again”. Further, when the cause information indicates that the position of the hand or finger is too far from the biological information acquisition unit 3, the guidance processing unit 12 reads a guidance message corresponding to the cause from the storage unit 4.
- the guidance message can be, for example, a message such as “Please bring your hand closer to the sensor and try again”.
- the guidance processing unit 12 displays the read guidance message on the display unit 2 to prompt the user to retry reading the vein pattern after moving the hand or finger to an appropriate position. Further, when the biometric authentication device 1 has a speaker, the guidance processing unit 12 may notify the user of a guidance message by voice through the speaker.
- the effective biological information extraction unit 13 extracts effective biological information that is information used for selecting registered biological information from the inappropriate input blood vessel image. To do.
- a part of the vein pattern 220 is also present in a part of the user's hand 210 included in the imaging region 200.
- the effective biological information extraction unit 13 can extract information related to the vein pattern from an inappropriate input blood vessel image obtained by photographing a hand placed at such an inappropriate position. Therefore, for example, the effective biological information extraction unit 13 uses, as effective biological information, feature points related to vein patterns existing in the hand region based on information indicating the hand region in the inappropriate input blood vessel image received from the processing unit 5. Extract.
- the effective biological information extraction unit 13 extracts, for example, branch points and end points of blood vessels as feature points. Therefore, the effective biological information extraction unit 13 binarizes the hand region of the inappropriate input blood vessel image using, for example, a local threshold method in order to extract the branch point and end point of the blood vessel from the inappropriate input blood vessel image. Next, the effective biological information extraction unit 13 performs a thinning process on the binarized hand region. Thereafter, the effective biological information extraction unit 13 scans the thinned hand region using a plurality of mask patterns to detect the position on the inappropriate input blood vessel image when it matches any mask pattern. To do. And the effective biological information extraction part 13 extracts the center pixel of the detected position as a feature point.
- the mask pattern is represented by, for example, 3 ⁇ 3 pixels, and has a binary pattern corresponding to a blood vessel branch point or end point.
- the effective biological information extraction unit 13 as other effective biological information, for example, as disclosed in Japanese Patent Application Laid-Open No. 2007-249339, from the hand region, the distribution of the curve direction of the blood vessel image and the blood vessel image You may extract the distribution of the direction of a trunk, or the frequency which shows the space
- the effective biological information extraction unit 13 extracts a blood vessel image by performing edge detection processing and thinning processing in the hand region. Then, the effective biological information extraction unit 13 approximates the extracted blood vessel image with a polygonal line.
- the effective biological information extraction unit 13 divides the blood vessel image approximated by the broken line into segment units including two line segments.
- the effective biological information extraction unit 13 calculates a direction vector of a perpendicular line from a connection point of the two line segments to a straight line connecting the end points of the two line segments as the bending direction. Then, the effective biological information extraction unit 13 uses the curve of the bending direction calculated for each of the plurality of segments as the distribution of the bending direction of the blood vessel image.
- the effective biological information extraction unit 13 can obtain a frequency indicating the distribution of the stem direction of the blood vessel image or the interval and the number of the stems in the blood vessel image, for example, by performing a fast Fourier transform on the hand region.
- the effective biological information extraction unit 13 may extract a frequency component related to the hand region image as effective biological information.
- the effective biological information extraction unit 13 is a reference in which the area of the hand region is an average area of the hand region when a human hand or finger is placed at an appropriate position with respect to the biological information acquisition unit 3.
- the inappropriate input blood vessel image is enlarged and corrected so as to coincide with the area.
- the reference area is determined in advance and stored in the storage unit 4. Then, the effective biological information extraction unit 13 performs frequency conversion processing such as fast Fourier transform or wavelet transform on the inappropriately input blood vessel image that has been enlarged and corrected.
- the effective biological information extraction unit 13 uses at least one frequency component in a predetermined frequency band of the image included in the inappropriate input blood vessel image obtained by the frequency conversion process as effective biological information.
- the predetermined frequency band can be a frequency band in which the frequency component changes when the contour shape or vein pattern of the hand region is different.
- the effective biological information extraction unit 13 performs edge detection processing on the inappropriate input blood vessel image or the enlarged input blood vessel image that has been subjected to enlargement correction, and creates an edge image in which the contour of the hand region and the vein are detected, Frequency conversion processing may be performed on the edge image.
- the effective biological information extracting unit 13 may use, as the effective biological information, a region having a predetermined shape circumscribing the hand region or a region having a predetermined shape inscribed in the hand region.
- the region having a predetermined shape can include a vein pattern in the hand region and a shape that does not include a portion unnecessary for the collation process other than the hand region as much as possible.
- the predetermined shape can be a rectangle or an ellipse.
- the effective biological information extraction unit 13 may extract different effective biological information depending on the cause of the inappropriate input blood vessel image.
- the valid biological information extraction unit 13 receives cause information indicating the cause of improperness from the processing unit 5.
- cause information indicates that the hand or finger is out of the imaging range of the biometric information acquisition unit 3 and a part of the biometric information to be used for verification is missing
- effective biometric information extraction is performed.
- the unit 13 extracts feature points in the hand region as effective biological information.
- the effective biological information extraction unit 13 extracts a frequency component related to the image of the hand region as effective biological information.
- the effective biological information extraction unit 13 passes the obtained effective biological information to the effective biological information matching unit 14. For each registered blood vessel image, the same processing as the processing of the effective biological information extraction unit 13 for the inappropriate blood vessel image is performed in advance. Then, effective biometric information is calculated for each registered blood vessel image.
- the effective biometric information calculated for each registered blood vessel image is stored in the storage unit 4 as one of the data related to the registered biometric information of the corresponding registered user. However, when the effective biometric information is the same as a feature value for verification described later, the feature amount for verification is used as effective biometric information related to the registered biometric information. Therefore, the memory
- the valid biometric information collating unit 14 receives the valid biometric information obtained from the inappropriate input blood vessel image received from the valid biometric information extracting unit 13, and the valid biometric information regarding each registered biometric information stored in the storage unit 4. Collate. Then, the effective biometric information matching unit 14 calculates the similarity between the effective biometric information obtained from the inappropriate input blood vessel image and the effective biometric information regarding each registered biometric information.
- the effective biometric information matching unit 14 determines a feature point located near the center of gravity of the hand region of the inappropriate input blood vessel image. 1 as a reference feature point.
- the valid biometric information matching unit 14 selects one of the feature points extracted from the registered blood vessel image of interest as the second reference feature point. Then, the valid biometric information matching unit 14 translates the registered blood vessel image so that the second reference feature point matches the first reference feature point. Thereafter, the valid biometric information matching unit 14 calculates the number of feature points of the registered blood vessel image that match the feature points of the inappropriate input blood vessel image while rotating the registered blood vessel image.
- the valid biometric information matching unit 14 repeats the above processing while changing the combination of the first reference feature point and the second reference feature point, and the feature of the registered blood vessel image that matches the feature point of the inappropriate input blood vessel image. Find the maximum number of points. Finally, the effective biometric information matching unit 14 obtains, as a similarity, a value obtained by dividing the maximum value of the number of matched feature points by the total number of feature points extracted from the input blood vessel image. Therefore, in this case, the similarity has a value of 0 to 1, and the higher the similarity between the vein pattern represented in the inappropriate input blood vessel image and the vein pattern represented in the registered blood vessel image, the higher the similarity value is 1. Get closer to.
- the valid biometric information matching unit 14 obtains each feature amount obtained from the inappropriate input blood vessel image and the registered blood vessel image to which attention is paid. The sum of the absolute values of the difference between the corresponding feature values obtained from the above is calculated.
- the effective biometric information collating unit 14 is obtained from the inappropriate input blood vessel image for each of the distribution in the bending direction of the blood vessel image, the distribution in the direction of the stem of the blood vessel image, or the frequency indicating the interval and number of the stems in the blood vessel image.
- the absolute value of the difference between the obtained value and the value obtained from the registered blood vessel image of interest is obtained.
- the effective biometric information collation part 14 makes the value obtained by totaling the absolute value of the obtained difference the similarity.
- the valid biological information matching unit 14 obtains an absolute value of a difference between a frequency component in a predetermined frequency band obtained from the inappropriate input blood vessel image and a frequency component in a corresponding frequency band obtained from the registered blood vessel image of interest.
- the effective biometric information collation part 14 makes the value obtained by totaling the absolute value of the obtained difference the similarity. In this case, as the degree of similarity between the shape or vein pattern of the hand region represented in the inappropriate input blood vessel image and the shape or vein pattern of the hand region represented in the registered blood vessel image increases, the similarity value approaches 0. .
- the effective biometric information matching unit 14 is a region that is effective biometric information. Is used as a template, and the similarity is calculated by pattern matching with the registered blood vessel image of interest. In this case, the effective biometric information matching unit 14 changes the relative position between the template and the registered blood vessel image, and calculates the correlation value c (i, j) between the template and the registered blood vessel image using the following formula. calculate.
- I (x, y) represents the pixel value of the pixel of the horizontal coordinate x and the vertical coordinate y included in the registered blood vessel image.
- T (xi, yj) represents the pixel value of the pixel of the horizontal coordinate (xi) and the vertical coordinate (yj) included in the template.
- I av is the average pixel value of the pixels included in the registered blood vessel image
- T av is the average pixel value of the pixels included in the template.
- i and j represent the amounts of deviation between the registered blood vessel image and the template in the horizontal and vertical directions, respectively.
- c (i, j) represents a correlation value when the registered blood vessel image is shifted by i pixels in the horizontal direction and j pixels in the vertical direction with respect to the template.
- the correlation value c (i, j) can take a value included between ⁇ 1 and 1.
- the correlation value c (i, j) is 1.
- the correlation value c (i, j) is -1.
- the valid biometric information collation unit 14 passes the similarity calculated for each registered biometric information to the selection unit 15 together with the registered user identification information regarding the registered biometric information corresponding to the similarity.
- the selection unit 15 selects registered biometric information similar to the input biometric information represented in the inappropriate input blood vessel image. Therefore, the selection unit 15 determines whether or not the similarity calculated for each registered biometric information satisfies a predetermined selection condition. When the similarity satisfies the selection condition, the selection unit 15 selects the registered biometric information corresponding to the similarity for use in the matching process with the appropriate input blood vessel image reacquired by the biometric information acquisition unit 3. .
- the selection condition is a value related to the maximum value of the number of feature points of the registered blood vessel image whose similarity matches the feature point of the inappropriate input blood vessel image, or a correlation value obtained by pattern matching For example, the similarity is equal to or higher than the selection threshold.
- the selection threshold is a value that prevents the selection biometric information corresponding to the specific registered user from being selected by the selection unit 15 when it is determined that there is no possibility that the user is a specific registered user. It is preferable to set to.
- the selection threshold value is set to the same value as a collation threshold value defined with respect to a condition for determining that the user authentication is successful, which will be described later.
- the selection threshold may be set to a value that eases the condition for selecting the registered biometric information, rather than the condition for the authentication determination unit 18 to determine that the authentication has succeeded.
- the selection unit 15 refers to the registered biometric information corresponding to the similarity determined to satisfy the predetermined condition by referring to the identification information of the registered user corresponding to the registered biometric information received from the valid biometric information matching unit 14.
- the data is associated with a selection flag indicating that it has been selected.
- the selection unit 15 stores the selection flag in the storage unit 4.
- the selection part 15 may replicate the data regarding the selected registration biometric information to the storage area different from the storage area where the data regarding all the registration biometric information are memorize
- the selection unit 15 may store the data related to the selected registered biometric information in a storage area accessible by the processing unit 5 faster than the storage area in which the data related to all registered biometric information is stored. preferable.
- the process part 5 can speed up the collation process using the data regarding the selected registration biometric information.
- the feature amount extraction unit 16 extracts a feature amount for the matching process from the appropriate input blood vessel image when the matching process is executed.
- the feature amount extraction unit 16 extracts feature points such as branch points and end points of blood vessels as feature amounts from the hand region, as in the case of the effective biological information extraction unit 13.
- the feature amount extraction unit 16 extracts, from the hand region, the distribution of the blood vessel image in the bending direction and the distribution of the stem direction of the blood vessel image or the frequency indicating the interval and number of stems in the blood vessel image as the feature amount.
- the feature quantity extraction unit 16 can extract the feature quantity using a method similar to the method described for the effective biological information extraction unit 13. Therefore, detailed description of the feature point extraction method is omitted here.
- the feature amount extracted by the feature amount extraction unit 16 may be the same type of feature amount as the feature amount extracted by the effective biological information extraction unit 13 or may be a different type of feature amount.
- the effective biological information extracting unit 13 extracts feature points such as branch points and end points of blood vessels as feature amounts, while the feature amount extracting unit 16 distributes the distribution direction of the blood vessel image and the distribution of the trunk direction of the blood vessel image. Or you may extract the frequency which shows the space
- both the effective biological information extraction unit 13 and the feature amount extraction unit 16 may extract feature points such as branch points and end points of blood vessels as feature amounts.
- the processing unit 5 can collate the input biometric information and the registered biometric information using different types of information. Can be improved.
- the feature amount extraction unit 16 passes feature amounts such as a plurality of feature points extracted from the appropriate input blood vessel image to the matching unit 17. For each registered blood vessel image, the same processing as the processing of the feature amount extraction unit 16 for the appropriate blood vessel image is performed in advance. A feature amount is also calculated for each registered blood vessel image. The feature amount calculated for each registered blood vessel image is stored in the storage unit 4 as one of the data related to the corresponding registered biometric information. Further, as will be described later, when the matching unit 17 performs a matching process using a matching method that does not use a feature quantity, for example, pattern matching, the feature quantity extraction unit 16 may be omitted.
- the collation unit 17 collates the vein pattern that is the input biometric information represented in the appropriate input blood vessel image with the vein pattern that is the preregistered biometric information. And the collation part 17 calculates
- the collation unit 17 can calculate the similarity by examining the number of feature points that match the feature points extracted from the registered blood vessel image among the feature points extracted from the proper input blood vessel image.
- the collation part 17 can calculate a similarity by calculating
- the collation unit 17 can obtain the similarity by pattern matching between the region inscribed or circumscribed with the hand region included in the appropriate input blood vessel image and the registered blood vessel image.
- the collation unit 17 can execute the collation process using a method similar to the method described for the effective biometric information collation unit 14. Therefore, detailed description of the collation process is omitted here.
- the collation part 17 calculates
- the feature amount extraction unit 16 and the collation unit 17 select the registered biometric information. Only the target of collation processing. That is, the feature quantity extraction unit 16 and the collation unit 17 use only data related to registered biometric information associated with the selection flag for the collation process. Thereby, the feature amount extraction unit 16 and the collation unit 17 perform the collation processing after re-inputting the input blood vessel image when the input blood vessel image once input is inappropriate and the input blood vessel image is acquired again. Can be reduced. Therefore, the processing unit 5 can shorten the time required for the entire biometric authentication process.
- the authentication determination unit 18 determines whether the highest similarity calculated by the verification unit 17 satisfies a predetermined authentication condition.
- the authentication condition is that the maximum similarity is a value related to the maximum value of the number of feature points of the registered blood vessel image that matches the feature point of the proper input blood vessel image as described above, or a correlation value obtained by pattern matching For example, the highest similarity is equal to or higher than the authentication determination threshold value. Further, when the highest similarity is the sum of absolute values of the difference between the feature amount extracted from the appropriate input blood vessel image and the feature amount extracted from the registered blood vessel image as described above, the authentication condition is, for example, the highest similarity. The degree is equal to or less than the authentication determination threshold.
- the authentication determination unit 18 determines that the input biometric information and the registered biometric information match when the highest similarity satisfies the authentication condition. Then, the authentication determination unit 18 authenticates the user as a registered user corresponding to the registered biometric information determined to match the input biometric information. When authenticating the user, the authentication determination unit 18 notifies the processing unit 5 of the authentication result. Then, the processing unit 5 permits an authenticated user to use a device in which the biometric authentication device 1 is mounted or a device to which the biometric authentication device 1 is connected.
- the authentication determination unit 18 determines that the input biometric information does not match the registered biometric information. In this case, the authentication determination unit 18 does not authenticate the user.
- the authentication determination unit 18 notifies the processing unit 5 of an authentication result indicating that the user authentication has failed.
- the process part 5 refuses that the user who was not authenticated uses the apparatus by which the biometrics apparatus 1 was mounted, or the apparatus to which the biometrics apparatus 1 was connected. Further, the processing unit 5 causes the display unit 2 to display a message indicating that the authentication has failed.
- the authentication determination threshold is preferably set to a value such that the authentication determination unit 18 succeeds in authentication only when the registered user is a user.
- the authentication determination threshold is preferably set to a value that causes the authentication determination unit 18 to fail authentication when another person different from the registered user is the user.
- the authentication determination threshold value may be a value obtained by adding a value obtained by multiplying the difference between the maximum value and the minimum value of similarity by 0.7 to the minimum value of similarity.
- FIG. 5 shows an operation flowchart of a biometric authentication process controlled by a computer program executed on the processing unit 5.
- the processing unit 5 acquires an input blood vessel image representing a vein pattern of the user's hand or finger, which is the user's biological information, from the biological information acquisition unit 3 (Step S ⁇ b> 101).
- the quality determination unit 11 of the processing unit 5 determines whether or not the input blood vessel image is appropriate for use in the matching process (step S102).
- the quality determination unit 11 can determine whether or not the input blood vessel image is appropriate based on, for example, the size of the hand region or the center of gravity extracted from the input blood vessel image.
- the pass / fail judgment unit 11 identifies the cause of the inappropriate input blood vessel image (step S102). S103). Then, the pass / fail determination unit 11 passes the cause information indicating the cause of the inappropriate input blood vessel image to the processing unit 5. The quality determination unit 11 also passes to the processing unit 5 information indicating the hand region in the inappropriate input blood vessel image that was used for the quality determination.
- the guidance processing unit 12 of the processing unit 5 reads a guide message corresponding to the cause of the inappropriateness from the storage unit 4 according to the cause information. Then, the guidance processing unit 12 displays the read guide message on the display unit 2 (step S104).
- the effective biological information extraction unit 13 of the processing unit 5 extracts effective biological information from the inappropriate input blood vessel image (step S105). Then, the effective biological information extraction unit 13 passes the extracted effective biological information to the effective biological information matching unit 14 of the processing unit 5.
- the effective biometric information collating unit 14 collates the effective biometric information extracted from the improper input blood vessel image with the effective biometric information related to all registered biometric information stored in advance in the storage unit 4 to thereby determine the similarity. Calculate (step S106). Then, the valid biometric information matching unit 14 passes the similarity calculated for each registered biometric information to the selection unit 15 of the processing unit 5 together with the registered user identification information regarding the registered biometric information corresponding to the similarity.
- the selection unit 15 selects registered biometric information corresponding to the similarity satisfying the selection condition (step S107). And the selection part 15 memorize
- the feature amount extraction unit 16 of the processing unit 5 extracts the collation feature amount from the proper input blood vessel image. Extract (step S108). Then, the feature quantity extraction unit 16 passes the extracted feature quantity to the matching unit 17 of the processing unit 5.
- the collation unit 17 determines whether there is registered biometric information selected by the selection unit 15 (step S109). If there is the selected registered biometric information (step S109—Yes), the collation unit 17 collates the collation feature quantity extracted from the appropriate input blood vessel image with the collation feature quantity regarding the selected registered biometric information.
- the matching unit 17 calculates the similarity between the input biometric information represented in the proper input blood vessel image and each selected biometric information (step S110). On the other hand, if there is no selected registered biometric information (step S109-No), the collation unit 17 collates the collation feature quantity extracted from the appropriate input blood vessel image with the collation feature quantity regarding all the registered biometric information. And the collation part 17 calculates the similarity of the input biometric information represented by the appropriate input blood vessel image, and all the registration biometric information, respectively (step S111).
- the collation unit 17 determines the highest similarity indicating that the input biometric information and the registered biometric information are the most similar among the calculated similarities (step S112). Then, the collation unit 18 passes the highest similarity to the authentication determination unit 18 of the processing unit 5 together with the identification information of the registered user regarding the registered biometric information corresponding to the highest similarity.
- the authentication determination unit 18 determines whether or not the highest similarity satisfies the authentication condition (step S113). If the highest similarity satisfies the authentication condition (step S113—Yes), the authentication determination unit 18 authenticates the user as a registered user corresponding to the highest similarity (step S114). On the other hand, when the highest similarity does not satisfy the authentication condition (No at Step S113), the authentication determination unit 18 does not authenticate the user (Step S115). After step S114 or S115, the processing unit 5 ends the biometric authentication process.
- the processing unit 5 may execute the processes in steps S105 to S107 while the biological information acquisition unit 3 generates the input blood vessel image again after the processes in steps S103 and S104.
- the processing unit 5 may execute the processes in steps S103 and S104 and the processes in steps S105 to S107 in parallel.
- the processing unit 5 may exchange the order of the processes in steps S103 and S104 and the processes in steps S105 to S107.
- the processing unit 5 may omit steps S105 to S107 when sufficient effective biometric information for selecting registered biometric information cannot be obtained from the inappropriate input blood vessel image.
- the processing unit 5 sets the area of the hand region in the inappropriate input blood vessel image obtained by the pass / fail determination unit 11 to a position where the human hand or finger is in an appropriate position with respect to the biological information acquisition unit 3. It is compared with a reference area which is an average area of the hand region on the input blood vessel image when placed. Then, when the ratio of the area of the hand region to the reference area is less than the area ratio necessary for extracting effective biological information, the processing unit 5 omits the processes of steps S105 to S107.
- the minimum area ratio necessary for extraction of effective biological information can be any value within the range of 0.3 to 0.5, for example.
- the processing unit 5 may authenticate the user as a user corresponding to the selected registered biometric information. In this case, the processing unit 5 can accurately authenticate the user because there is no other registered biometric information that may match the input biometric information without acquiring a proper input blood vessel image again. is there. In this case, the processes of steps S110 to S113 and S115 may be omitted. Therefore, the processing unit 5 can shorten the time required for the entire biometric authentication process.
- the matching unit 17 does not use the feature amount extracted from the blood vessel image, and thus the process of step S108 may be omitted.
- the collation unit 17 executes pattern matching between the appropriate input blood vessel image and the selected registered blood vessel image or all the registered blood vessel images.
- FIG. 6 is a conceptual diagram showing a comparison between the time required to execute the biometric authentication process according to one embodiment and the time required to execute the biometric authentication process according to the conventional technique.
- the horizontal axis represents elapsed time.
- the block column 600 represents the time of biometric authentication processing when the input blood vessel image obtained first is appropriate.
- the block row 610 is a case where the conventional biometric authentication device executes verification processing for all registered biometric information after re-acquiring the input blood vessel image when the first obtained input blood vessel image is inappropriate. Represents the time of biometric authentication processing.
- the block sequence 620 selects the registered biometric information to be subjected to the matching process based on the inappropriate input blood vessel image when the first input blood vessel image obtained by the biometric authentication device 1 is inappropriate. It represents the time for biometric authentication processing.
- a block 601 represents the time required to acquire the input blood vessel image.
- a block 602 represents the time required to determine whether the acquired input blood vessel image is acceptable.
- a block 603 represents the time required to extract the feature amount related to the vein pattern from the input blood vessel image.
- block 604 represents the time required for the matching process for all registered biometric information.
- a block 605 represents the time required for the matching process for only the registered biometric information selected based on the inappropriate input blood vessel image.
- block 606 represents the time required to extract valid biometric information regarding the vein pattern or hand region from an inappropriate input blood vessel image.
- Block 607 represents the time required to select some of the registered biometric information from the registered biometric information based on the inappropriate input blood vessel image.
- a block 608 represents the time required for the authentication determination process.
- both the biometric authentication device according to the conventional technique and the biometric authentication device 1 according to the present embodiment are checked against all registered biometric information. Process. Therefore, the time required for the biometric authentication device according to the prior art to execute the entire biometric authentication process is equal to the time required for the biometric authentication device 1 according to the present embodiment to execute the entire biometric authentication process.
- the biometric authentication device when the input blood vessel image obtained first is inappropriate, the biometric authentication device according to the conventional technique re-acquires the input blood vessel image, and ends the feature amount extraction processing for the re-acquired input blood vessel image.
- the verification process cannot be executed.
- the biometric authentication process according to the conventional technique after the feature amount extraction process is completed, a matching process must be performed on all registered biometric information.
- the processing load on the processing unit 5 for the reacquisition process is small.
- the processing unit 5 is based on the inappropriate input blood vessel image.
- a process of selecting a part of registered biometric information from the registered biometric information can be executed.
- the biometric authentication device 1 may perform the matching process only on the selected registered biometric information after the feature amount extraction process on the re-acquired input blood vessel image is completed. Therefore, the biometric authentication device 1 can reduce the time required for the entire biometric authentication process as compared with the biometric authentication device according to the prior art by the difference between the verification processing time indicated by the block 604 and the verification processing time indicated by the block 605.
- the biometric authentication device provides a response to the user when a part of the biometric information represented in the input biometric image is missing or the biometric information is unclear.
- this biometric authentication device selects registered biometric information similar to the incomplete input biometric information from all registered biometric information registered in advance. And this biometrics authentication apparatus collates only the input biometric information represented by the appropriate input biometric image obtained when the user made the biometric information acquisition part read again by biometric information, and the selected registration biometric information. .
- this biometric authentication device can suppress an increase in processing time required for executing the entire biometric authentication process by reducing the number of registered biometric information for executing the biometric authentication process after re-inputting the input biometric image. .
- this invention is not limited to said embodiment.
- the storage unit uses the valid biometric information extracted from the improper input blood vessel image at the previous acquisition as the previous valid biometric information. You may remember it.
- the effective biological information matching unit of the processing unit uses the effective biological information acquired from the newly acquired input blood vessel image. The latest valid biometric information is collated with the previous valid biometric information. Then, the valid biometric information matching unit calculates the similarity between the previous valid biometric information and the latest valid biometric information.
- the selection unit of the processing unit is already selected when it is determined that the similarity between the previous valid biometric information and the latest valid biometric information does not satisfy a predetermined selection condition, and the latest valid biometric information does not match the previous valid biometric information. All the registered biometric information is not selected. For example, when the selection flag is associated with the data related to the selected registered biometric information, the selection unit deletes the selection flag. Alternatively, when data related to the selected registered biometric information is stored in a storage area separate from the storage area in which other registered biometric information is stored, the selection unit selects the registration selected from the separate storage area Delete data related to biometric information.
- the effective biometric information matching unit of the processing unit compares the valid biometric information related to each of all the registered biometric information with the latest valid biometric information. The similarity is calculated respectively.
- the selection unit again selects the registered biometric information corresponding to the degree of similarity that satisfies the selection condition.
- the selection unit may maintain the previously selected registered biometric information as it is.
- the valid biometric information collating unit collates the valid biometric information related to each of the registered biometric information not selected and the latest valid biometric information.
- a selection part makes all the registration biometric information already selected unselected, and selects the registration biometric information corresponding to the similarity which satisfy
- the valid biometric information collation unit and the selection unit reselect the registered biometric information by collating the valid biometric information related to each of all the registered biometric information and the latest valid biometric information.
- the processing unit when any error occurs in the processing of each unit included in the processing unit, or when a command for clearing the selected registered biometric information is input via the input unit, all already selected The registered biometric information may be unselected.
- the authentication determination unit of the processing unit may determine whether or not the similarity satisfies the authentication condition every time the collation unit calculates the similarity corresponding to any registered biometric information. In this case, the authentication determination unit recognizes the user as a registered user corresponding to the similarity that is first determined to satisfy the authentication condition. According to this modification, the processing unit may not have to collate the input biometric information with all of the selected registered biometric information, and thus may further reduce the time required for the entire biometric authentication process. is there.
- the biometric authentication device acquires user identification information via the input unit, so that only registered biometric information specified by the identification information is obtained. May be matched.
- the biometric authentication device has subsequently obtained a verification result between the effective biometric information extracted from the input blood vessel image inappropriate for the verification process and the effective biometric information regarding the specified registered biometric information. You may utilize in order to assist the biometrics authentication process based on collation with the input biometric information represented by the appropriate input blood vessel image, and registration biometric information.
- the biometric authentication device when the quality determination unit of the processing unit determines that the input blood vessel image is inappropriate for use in the verification process, the effective biometric information extraction unit and the effective biometric information verification of the processing unit The process by the part is executed.
- the biometric authentication device obtains the similarity between the effective biometric information extracted from the inappropriate input blood vessel image and the effective biometric information related to the registered biometric information specified by the user identification information.
- the collation unit of the processing unit calculates the similarity by collating the input biometric information represented in the appropriate input blood vessel image obtained thereafter with the registered biometric information specified by the user identification information. To do.
- the authentication determination unit of the processing unit when the similarity calculated by the verification unit satisfies the authentication condition described above and the similarity calculated by the effective biometric information verification unit satisfies the selection condition described above. May be authenticated.
- the biometric authentication device can perform biometric authentication processing using a plurality of input blood vessel images photographed under different conditions, thereby improving the authentication accuracy.
- the biometric authentication apparatus performs a verification process using the inappropriate input blood vessel image while the input blood vessel image is reacquired. Can be executed. Therefore, this biometric authentication apparatus can prevent an increase in the time required for the biometric authentication process even though a plurality of input blood vessel images are used for the biometric authentication process.
- the biometric authentication device and the biometric authentication method disclosed in the present specification execute a biometric authentication process between the biometric information of the user and the preregistered biometric information in order for the user to perform some operation.
- an apparatus or system includes a computer system in which one or more terminals and a server are connected via a communication network, or an access control system.
- each terminal is provided with a biological information acquisition unit, and the biological image acquired by the biological information acquisition unit is transmitted to the server.
- a server performs the registration process or biometrics authentication process of a biometric image by performing the function of the process part of said embodiment.
- the processing unit of the portable memory device in which the biological information acquisition unit, the storage unit, the processing unit, and the data transmission interface conforming to the standards such as Universal Serial Bus are integrally formed in the above embodiment. You may have each function of the process part of a computer.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
Description
例えば、生体認証装置が、生体情報を表す生体画像として利用者の手の静脈パターンが撮像された画像を取得する場合、センサの撮像範囲内に手全体が含まれていなければ、得られた画像には手の静脈パターンの一部しか写らなくなってしまう。あるいは、センサに対する理想的な手の位置よりも遠くに手が翳されると、得られた画像に表された手の静脈パターンの像は小さく、その結果、画像上でその静脈パターンの細部が不鮮明となってしまう。
そこで、生体画像が適正か否か判断し、適正でないと判断した場合には、利用者に対して生体画像を再入力させる技術が開発されている。
上記の一般的な記述及び下記の詳細な記述の何れも、例示的かつ説明的なものであり、請求項のように、本発明を限定するものではないことを理解されたい。
この生体認証装置は、利用者に対する生体認証処理を実行する際、利用者の生体情報を表した画像である入力生体画像を取得する。そしてこの生体認証装置は、入力生体画像に表された利用者の生体情報である入力生体情報を、生体認証装置に予め登録された複数の登録利用者の生体情報である登録生体情報のそれぞれと照合する。この生体認証装置は、照合処理の結果に基づいて、入力生体情報と何れかの登録生体情報が一致すると判定した場合、利用者を、入力生体情報と一致すると判定された登録生体情報に対応する登録利用者として認証する。
ここで、この生体認証装置は、入力生体画像に表された入力生体情報が不鮮明であったり、入力生体画像に照合に使用されるべき生体情報の一部しか写っていないなど、入力生体情報が不完全である場合、利用者に対して生体情報の再入力を指示する。一方で、この生体認証装置は、予め登録されている全ての登録生体画像から、その不完全な入力生体情報と類似する登録生体情報が表された登録生体画像を選択する。そしてこの生体認証装置は、利用者が生体情報を再度読み取らせることにより得られた適正な入力生体画像に表された入力生体情報と、選択された登録生体画像に表された登録生体情報のみを照合する。これにより、この生体認証装置は、入力生体画像の再入力後に生体認証処理を実行する登録生体画像の数を減らすことで、生体認証処理全体の実行に要する処理時間を短縮する。
また、本明細書において、「照合処理」という用語は、入力生体情報と登録生体情報の類似度合いを表す類似度を算出する処理を示すために使用される。また、「生体認証処理」という用語は、照合処理だけでなく、照合処理により求められた類似度を用いて、利用者を認証するか否かを決定する処理を含む、認証処理全体を示すために使用される。
なお、生体認証装置1は、例えば、キーボード、マウス、またはタッチパッドなどの入力部6を有してもよい。そして生体認証装置1は、入力部6を介して利用者により入力されたコマンド、データ、あるいは利用者の識別情報を取得し、そのコマンド、データあるいは利用者の識別情報を処理部5へ渡してもよい。ただし、利用者が生体情報以外の情報を生体認証装置1に対して入力する必要がない場合、この入力部6は省略されてもよい。
なお、表示部2と生体情報取得部3は、一体的に形成されていてもよい。
さらに記憶部4は、各登録生体情報に関するデータの一つとして、照合処理用の特徴量とは別個に、登録血管画像から抽出された、後述する選択処理用の有効生体情報を記憶してもよい。
また記憶部4は、生体情報取得部3が適正な血管画像を取得可能な位置へ、手または指を配置させるためのガイダンスメッセージを記憶する。
また、図2(b)では、撮影領域200に含まれる利用者の手210の範囲が、図2(a)よりも広くなっているものの、まだ親指近傍の静脈パターンが撮影領域200から外れている。そのため、処理部5は、図2(b)に示された状態で取得された入力血管画像を用いて照合処理を行うと、図2(a)に示された状態で取得された入力血管画像を用いるよりも照合精度の向上を期待できる。しかし処理部5は、利用者の静脈パターンを登録利用者の静脈パターンと正確に照合できない可能性がある。
これに対し、図2(c)では、撮影領域200内に、利用者の手210の略全域が含まれている。そのため、図2(c)に示された状態で取得された入力血管画像には、手のひら部分の静脈パターンの全部が写る。この結果、処理部5は、図2(c)に示された状態で取得された入力血管画像を用いて照合処理を行うことにより、利用者の静脈パターンを登録利用者の静脈パターンと正確に照合することができる。
また、図3(b)では、撮影領域300に占める利用者の手310の面積が、図3(a)よりも大きくなっているものの、手310の面積は、利用者の手が適正な位置に置かれた場合の手の面積よりも小さい。そのため、処理部5は、図3(b)に示された状態で取得された入力血管画像を用いて照合処理を行うと、図3(a)に示された状態で取得された入力血管画像を用いるよりも照合精度の向上を期待できる。しかし処理部5は、利用者の静脈パターンを登録利用者の静脈パターンと正確に照合できない可能性がある。
これに対し、図3(c)では、利用者の手が適正な位置に置かれているため、撮影領域300内に占める利用者の手310の面積が十分に大きくなっている。そのため、図3(c)に示された状態で取得された入力血管画像では、手のひら部分の静脈パターンの微細構造も判別可能となっている。この結果、処理部5は、図3(c)に示された状態で取得された入力血管画像を用いて照合処理を行うことにより、利用者の静脈パターンを登録利用者の静脈パターンと正確に照合することができる。
例えば、良否判定部11は、入力血管画像の各画素の画素値を2値化閾値と比較することにより、入力血管画像を2値化する。2値化閾値は、例えば、入力血管画像の画素値の平均値とすることができる。そして、入力血管画像において、利用者の手が写っている領域に含まれる画素値が高く、何も写っていいない背景領域に含まれる画素値が低い場合、良否判定部11は、画素値が2値化閾値よりも高い場合、その画素を手領域候補画素とする。そして良否判定部11は、入力血管画像全体に対して得られた手領域候補画素に対してラベリング処理を行い、手領域候補画素が連結された領域を、照合に使用される静脈パターンを含む部位の像である手領域として検出する。
なお、利用者の手が写っている領域に含まれる画素の値が背景領域に含まれる画素の値より高くなるか低くなるかは、生体情報取得部3の構成によって決まる。そのため、利用者の手が写っている領域に含まれる画素の値が背景領域に含まれる画素の値より低くなる場合、良否判定部11は、2値化閾値よりも低い画素値を有する画素を手領域候補画素とすればよい。
以下では、不適正であると判定された入力血管画像を不適正入力血管画像と呼び、一方、適正であると判定された入力血管画像を適正入力血管画像と呼ぶ。
また、良否判定部11は、入力血管画像の左右の何れか一方の画像端に接触している手領域内の画素数と、他方の画像端に接触している手領域内の画素数との差が所定の閾値以上である場合、入力血管画像は不適正であると判定してもよい。この所定の閾値は、例えば、入力血管画像上の手領域の上下方向の長さの1/3あるいは1/4とすることができる。
さらに、手領域の重心位置及び画像端と接触する手領域内の画素数が、入力血管画像を不適正と判定する基準を満たさず、かつ手領域の面積が小さいために入力血管画像が不適正と判定されることもある。この場合、良否判定部11は、不適正となった原因は手または指の位置が生体情報取得部3に対して遠すぎることであると推定する。
そこで、有効生体情報抽出部13は、例えば、処理部5から受け取った不適正入力血管画像中の手領域を示す情報に基づいて、手領域内に存在する静脈パターンに関する特徴点を有効生体情報として抽出する。
そして有効生体情報抽出部13は、拡大補正された不適正入力血管画像に対して、高速フーリエ変換あるいはウェーブレット変換などの周波数変換処理を実行する。有効生体情報抽出部13は、周波数変換処理により得られた不適正入力血管画像に含まれる像の少なくとも一つの所定の周波数帯域の周波数成分を有効生体情報とする。なお、所定の周波数帯域は、例えば、手領域の輪郭形状あるいは静脈パターンが異なると周波数成分が変化する周波数帯域とすることができる。また有効生体情報抽出部13は、不適正入力血管画像あるいは拡大補正された不適正入力血管画像に対してエッジ検出処理を行って、手領域の輪郭及び静脈が検出されたエッジ画像を作成し、そのエッジ画像に対して周波数変換処理を行ってもよい。
なお、各登録血管画像に関しても、不適正血管画像に対する上記の有効生体情報抽出部13の処理と同様の処理が予め実施される。そして各登録血管画像について有効生体情報が算出される。各登録血管画像について算出された有効生体情報は、記憶部4に、対応する登録利用者の登録生体情報に関するデータの一つとして記憶される。ただし、有効生体情報が、後述する照合用の特徴量と同一である場合には、その照合用の特徴量が登録生体情報に関する有効生体情報として使用される。そのため、記憶部4は、登録生体情報に関する有効生体情報を、登録生体情報に関する照合用の特徴量と別個に記憶しなくてもよい。
最後に、有効生体情報照合部14は、一致した特徴点の個数の最大値を、入力血管画像から抽出された特徴点の総数で割った値を類似度として求める。したがって、この場合、類似度は0~1の値を持ち、不適正入力血管画像に表された静脈パターンと登録血管画像に表された静脈パターンの類似度合いが高いほど、類似度の値は1に近づく。
例えば、有効生体情報照合部14は、血管像の湾曲方向の分布、血管像の幹の方向の分布あるいは血管像における幹の間隔及び本数を示す周波数のそれぞれについて、不適正入力血管画像から得られた値と着目する登録血管画像から得られた値の差の絶対値を求める。そして有効生体情報照合部14は、得られた差の絶対値を合計することにより得られた値を類似度とする。
あるいは、有効生体情報照合部14は、不適正入力血管画像から得られた所定の周波数帯域の周波数成分と着目する登録血管画像から得られた対応する周波数帯域の周波数成分の差の絶対値を求める。そして有効生体情報照合部14は、得られた差の絶対値を合計することにより得られた値を類似度とする。
この場合、不適正入力血管画像に表された手領域の形状または静脈パターンと、登録血管画像に表された手領域の形状または静脈パターンの類似度合いが高いほど、類似度の値は0に近づく。
この場合、有効生体情報照合部14は、テンプレートと登録血管画像の相対的な位置を様々に変えつつ、下記の式を用いて、テンプレートと登録血管画像間の相関値c(i,j)を算出する。
選択条件は、上記のように、類似度が不適正入力血管画像の特徴点と一致する登録血管画像の特徴点の個数の最大値に関する値である場合、あるいは、パターンマッチングにより得られた相関値である場合、例えば、類似度が選択用閾値以上となることである。
また、選択条件は、上記のように、類似度が不適正入力血管画像から抽出された特徴量と登録血管画像から抽出された特徴量の差の絶対値和である場合、例えば、類似度が選択用閾値以下となることである。
選択用閾値は、利用者が特定の登録利用者である可能性が全くないと判断される場合に、その特定の登録利用者に対応する登録生体情報が選択部15により選択されなくなるような値に設定されることが好ましい。例えば、選択用閾値は、後述する、利用者の認証に成功したと判定する条件に関して規定される照合閾値と同一の値に設定される。あるいは、選択用閾値は、認証判定部18が認証に成功したと判定する条件よりも、登録生体情報が選択される条件が緩和されるような値に設定されてもよい。
なお、特徴量抽出部16が抽出する特徴量は、有効生体情報抽出部13が抽出する特徴量と同じ種類の特徴量であってもよく、あるいは異なる種類の特徴量であってもよい。例えば、有効生体情報抽出部13が血管の分岐点及び端点などの特徴点を特徴量として抽出する一方、特徴量抽出部16は血管像の湾曲方向の分布と、血管像の幹の方向の分布あるいは血管像における幹の間隔及び本数を示す周波数とを特徴量として抽出してもよい。また、有効生体情報抽出部13及び特徴量抽出部16の両方とも、血管の分岐点及び端点などの特徴点を特徴量として抽出してもよい。有効生体情報と照合用の特徴量が同一である場合、登録生体情報に関するデータとして、何れか一方のみが記憶されればよいので、記憶部4に記憶されるデータ量が少なくて済む。一方、有効生体情報と照合用の特徴量が異なる場合、処理部5は、異なる種類の情報を用いて入力生体情報と登録生体情報を照合できるので、単一種類の情報を用いるよりも認証精度を向上することができる。
なお、各登録血管画像に関しても、適正血管画像に対する上記の特徴量抽出部16の処理と同様の処理が予め実施される。そして各登録血管画像についても特徴量が算出される。各登録血管画像について算出された特徴量は、記憶部4に、対応する登録生体情報に関するデータの一つとして記憶される。
また、後述するように、照合部17が、特徴量を使用しない照合方法、例えば、パターンマッチングを用いて照合処理を実行する場合、特徴量抽出部16は省略されてもよい。
認証条件は、最高類似度が、上記のように、適正入力血管画像の特徴点と一致する登録血管画像の特徴点の個数の最大値に関する値である場合、またはパターンマッチングにより得られた相関値である場合、例えば、最高類似度が認証判定閾値以上となることである。また、認証条件は、最高類似度が、上記のように、適正入力血管画像から抽出された特徴量と登録血管画像から抽出された特徴量の差の絶対値和である場合、例えば、最高類似度が認証判定閾値以下となることである。
認証判定閾値は、登録利用者本人が利用者である場合にのみ、認証判定部18が認証に成功するような値に設定されることが好ましい。そして認証判定閾値は、登録利用者とは異なる他人が利用者である場合には、認証判定部18が認証に失敗するような値に設定されることが好ましい。例えば、認証判定閾値は、類似度の取りうる最大値と最小値の差に0.7を乗じた値を、類似度の最小値に加えた値とすることができる。
図5に示されるように、処理部5は、生体情報取得部3から、利用者の生体情報である利用者の手または指の静脈パターンを表す入力血管画像を取得する(ステップS101)。そして処理部5の良否判定部11は、入力血管画像が照合処理に利用するために適正か否か判定する(ステップS102)。良否判定部11は、例えば、入力血管画像から抽出した手領域のサイズまたは重心位置に基づいて、入力血管画像が適正か否かを判定できる。入力血管画像が照合処理に利用するには不適正な不適正入力血管画像である場合(ステップS102-No)、良否判定部11は、入力血管画像が不適正となった原因を特定する(ステップS103)。そして良否判定部11は、入力血管画像が不適正となった原因を示す原因情報を処理部5へ渡す。また良否判定部11は、良否判定に利用された、不適正入力血管画像中の手領域を示す情報も処理部5へ渡す。
選択部15は、選択条件を満たす類似度に対応する登録生体情報を選択する(ステップS107)。そして選択部15は、選択された登録生体情報に関するデータを識別する選択フラグを記憶部4に記憶するか、選択された登録生体情報に関するデータを全ての登録生体情報に関するデータと別個の記憶領域に複製する。
照合部17は、選択部15によって選択された登録生体情報があるか否か判定する(ステップS109)。選択された登録生体情報があれば(ステップS109-Yes)、照合部17は、適正入力血管画像から抽出された照合用特徴量と選択された登録生体情報に関する照合用特徴量を照合する。そして照合部17は、適正入力血管画像に表された入力生体情報と選択された各登録生体情報の類似度をそれぞれ算出する(ステップS110)。
一方、選択された登録生体情報がなければ(ステップS109-No)、照合部17は、適正入力血管画像から抽出された照合用特徴量と全ての登録生体情報に関する照合用特徴量を照合する。そして照合部17は、適正入力血管画像に表された入力生体情報と全ての登録生体情報の類似度をそれぞれ算出する(ステップS111)。
一方、最高類似度が認証条件を満たさない場合(ステップS113-No)、認証判定部18は利用者を認証しない(ステップS115)。
ステップS114またはS115の後、処理部5は、生体認証処理を終了する。
また、処理部5は、不適正入力血管画像から、登録生体情報を選択するために十分な有効生体情報が得られないときには、ステップS105~S107の処理を省略してもよい。例えばステップS102の後に、処理部5は、良否判定部11により求められた不適正入力血管画像に占める手領域の面積を、人の手または指が生体情報取得部3に対して適正な位置に置かれたときの入力血管画像上の手領域の平均的な面積である基準面積と比較する。そして処理部5は、基準面積に対する手領域の面積の比が、有効生体情報の抽出に最低限必要な面積比未満である場合、ステップS105~S107の処理を省略する。なお、有効生体情報の抽出に最低限必要な面積比は、例えば、0.3~0.5の範囲に含まれる何れかの値とすることができる。
また、各ブロック列において、ブロック601は、入力血管画像を取得するために要する時間を表す。またブロック602は、取得された入力血管画像の良否を判定するために要する時間を表す。さらにブロック603は、入力血管画像から静脈パターンに関する特徴量を抽出するために要する時間を表す。さらにブロック604は、全ての登録生体情報を対象とする照合処理に要する時間を表す。一方、ブロック605は、不適正な入力血管画像に基づいて選択された登録生体情報のみを対象とする照合処理に要する時間を表す。さらに、ブロック606は、不適正な入力血管画像から静脈パターンあるいは手領域に関する有効生体情報を抽出するために要する時間を表す。ブロック607は、不適正な入力血管画像に基づいて登録生体情報から一部の登録生体情報を選択するために要する時間を表す。そしてブロック608は、認証判定処理に要する時間を表す。
これに対し、生体情報取得部3による入力血管画像の再取得処理の実行中、その再取得処理に対する処理部5の処理負荷は少ない。そのため、生体認証装置1が入力血管画像を再取得し、再取得された入力血管画像に対して良否判定処理を実行している間に、処理部5は、不適正な入力血管画像に基づいて登録生体情報から一部の登録生体情報を選択する処理を実行できる。この結果、生体認証装置1は、再取得された入力血管画像に対する特徴量抽出処理が終了した後に、選択された登録生体情報のみを対象として照合処理を実行すればよい。従って、ブロック604に示される照合処理の時間とブロック605に示される照合処理の時間の差だけ、生体認証装置1は、従来技術による生体認証装置よりも生体認証処理全体に要する時間を短縮できる。
一方、選択部は、最新有効生体情報が前回有効生体情報と一致すると判定した場合、既選択の登録生体情報をそのまま維持してもよい。
さらに、既に選択されている登録生体情報が存在する場合において、処理部が生体情報取得部により生成された入力血管画像から手領域を検出できなかった場合も、既に選択されている全ての登録生体情報を未選択とする。この場合も、生体情報取得部に翳されていた手が外されたと想定されるので、生体認証装置に対して生体認証しようとする利用者が、異なる利用者に代わった可能性があるためである。そして有効生体情報照合部及び選択部は、全ての登録生体情報のそれぞれに関する有効生体情報と、最新有効生体情報を照合することにより、登録生体情報を再選択する。
また処理部は、処理部が有する各部の処理において何らかのエラーが発生した場合、あるいは、選択された登録生体情報をクリアするコマンドが入力部を介して入力された場合も、既に選択されている全ての登録生体情報を未選択としてもよい。
2 表示部
3 生体情報取得部
4 記憶部
5 処理部
6 入力部
11 良否判定部
12 ガイダンス処理部
13 有効生体情報抽出部
14 有効生体情報照合部
15 選択部
16 特徴量抽出部
17 照合部
18 認証判定部
Claims (10)
- 利用者の生体情報を取得して、該生体情報を表す第1の入力生体画像を生成する生体情報取得部と、
予め登録された少なくとも一人の登録利用者の登録生体情報に関するデータを記憶する記憶部と、
処理部であって、
前記第1の入力生体画像に表された生体情報を前記登録生体情報との照合に使用するために、前記第1の入力生体画像が適性か否かを判定する良否判定機能と、
前記第1の入力生体画像が不適正であると判定された場合、前記登録生体情報のうち、前記第1の入力生体画像に表された生体情報と類似する登録生体情報を選択する選択機能と、
前記生体情報取得部が利用者の生体情報を再取得することにより生成された第2の入力生体画像に表された生体情報と選択された前記登録生体情報を照合する照合処理機能と、
を実現する処理部と、
を有する生体認証装置。 - 前記処理部は、前記生体情報取得部が前記第2の入力生体画像を生成している間に、前記選択機能により前記第1の入力生体画像に表された生体情報と類似する登録生体情報を選択する、請求項1に記載の生体認証装置。
- 前記良否判定機能は、前記第1の入力生体画像から照合に使用される生体情報を含む利用者の部位の像を検出し、前記第1の入力生体画像における該像の位置に基づいて、前記照合に使用される生体情報の少なくとも一部が前記第1の入力生体画像において欠落していると判定されるとき、前記第1の入力生体画像は不適正であると判定する、請求項1または2に記載の生体認証装置。
- 前記良否判定機能は、前記第1の入力生体画像から、照合に使用される生体情報を含む利用者の部位の像を検出し、該像が前記第1の入力生体画像に占める面積が所定の最小許容面積以下である場合、前記第1の入力生体画像は不適正であると判定する、請求項1または2に記載の生体認証装置。
- 前記処理部は、前記選択された登録生体情報が一つのみである場合、利用者を当該登録生体情報に対応する登録利用者として認証する、請求項1~4の何れか一項に記載の生体認証装置。
- 前記処理部は、
前記第1の入力生体画像が不適正であると判定された場合、当該第1の入力生体画像から、利用者の生体情報に関する特徴量である有効生体情報を抽出する有効生体情報抽出機能と、
前記第1の入力生体画像が不適正であると判定された場合、前記有効生体情報を前記登録生体情報のそれぞれに関するデータと照合することにより、前記登録生体情報のそれぞれと、前記第1の入力生体画像に表された生体情報との類似度を算出する有効生体情報照合機能と
をさらに実現し、
前記選択機能は、前記登録生体情報のうち、所定の選択条件を満たす前記類似度に対応する登録生体情報を選択する、請求項1~5の何れか一項に記載の生体認証装置。 - 前記有効生体情報抽出機能は、前記第1の入力生体画像から抽出された前記有効生体情報を前回有効生体情報として前記記憶部に記憶するとともに、前記第1の入力生体画像が生成された後に、前記生体情報取得部により生成された第3の入力生体画像が前記良否判定機能により不適正であると判定された場合、該第3の入力生体画像から前記有効生体情報を抽出し、
前記選択機能は、前記第3の入力生体画像から抽出された前記有効生体情報と前記前回有効生体情報とが一致するか否か判定し、前記有効生体情報と前記前回有効生体情報とが一致しないと判定した場合、選択された登録生体情報を未選択とする、請求項6に記載の生体認証装置。 - 利用者の生体情報を取得して、該生体情報を表す第1の入力生体画像を生成し、
前記第1の入力生体画像に表された生体情報を記憶部に記憶された少なくとも一人の登録利用者の登録生体情報との照合に使用するために、前記第1の入力生体画像が適性か否かを判定し、
前記第1の入力生体画像が不適正であると判定された場合、前記登録生体情報のうち、前記第1の入力生体画像に表された生体情報と類似する登録生体情報を選択し、
利用者の生体情報を再取得することにより第2の入力生体画像を生成し、
前記第2の入力生体画像に表された生体情報と選択された前記登録生体情報を照合する、
ことを含む生体認証方法。 - 生体情報取得部により生成された第1の入力生体画像に表された利用者の生体情報を、記憶部に記憶された少なくとも一人の登録利用者の登録生体情報との照合に使用するために、前記第1の入力生体画像が適性か否かを判定し、
前記第1の入力生体画像が不適正であると判定された場合、前記登録生体情報のうち、前記第1の入力生体画像に表された生体情報と類似する登録生体情報を選択し、
前記生体情報取得部により生成された第2の入力生体画像に表された生体情報と選択された前記登録生体情報を照合する、
ことをコンピュータに実行させる生体認証用コンピュータプログラム。 - 利用者の生体情報を取得して、該生体情報を表す第1の入力生体画像を生成する生体情報取得部と、
予め登録された少なくとも一人の登録利用者の登録生体情報に関するデータを登録利用者の識別情報とともに記憶する記憶部と、
利用者の識別情報を取得する入力部と、
処理部であって、
前記第1の入力生体画像に表された生体情報を前記登録生体情報との照合に使用するために、前記第1の入力生体画像が適性か否かを判定する良否判定機能と、
前記第1の入力生体画像が不適正であると判定された場合、前記入力部を介して入力された利用者の識別情報により特定された登録生体情報と、前記第1の入力生体画像に表された生体情報との第1の類似度を算出する有効生体情報照合機能と、
前記生体情報取得部が利用者の生体情報を再取得することにより生成された第2の入力生体画像に表された生体情報と前記利用者の識別情報により特定された登録生体情報を照合することにより、第2の類似度を算出する照合処理機能と、
前記第1の類似度及び前記第2の類似度がそれぞれ所定の認証条件を満たす場合、利用者を前記利用者の識別情報により特定された登録生体情報に対応する登録利用者として認証する認証機能と、
を実現する処理部と、
を有する生体認証装置。
Priority Applications (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011519354A JP5445584B2 (ja) | 2009-06-17 | 2009-06-17 | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム |
EP09846168.4A EP2444933B1 (en) | 2009-06-17 | 2009-06-17 | Biometric authentication device, biometric authentication method and computer program for biometric authentication |
PCT/JP2009/061037 WO2010146675A1 (ja) | 2009-06-17 | 2009-06-17 | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム |
KR1020117029711A KR101328358B1 (ko) | 2009-06-17 | 2009-06-17 | 생체 인증 장치, 생체 인증 방법 및 생체 인증용 컴퓨터 프로그램이 기록된 컴퓨터 판독가능한 기록매체 |
CN200980159911.6A CN102460507B (zh) | 2009-06-17 | 2009-06-17 | 生物体认证装置、生物体认证方法以及生物体认证用计算机程序 |
US13/326,035 US8565494B2 (en) | 2009-06-17 | 2011-12-14 | Biometric authentication device, biometric authentication method, and computer program for biometric authentication |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2009/061037 WO2010146675A1 (ja) | 2009-06-17 | 2009-06-17 | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US13/326,035 Continuation US8565494B2 (en) | 2009-06-17 | 2011-12-14 | Biometric authentication device, biometric authentication method, and computer program for biometric authentication |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2010146675A1 true WO2010146675A1 (ja) | 2010-12-23 |
Family
ID=43356013
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2009/061037 WO2010146675A1 (ja) | 2009-06-17 | 2009-06-17 | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム |
Country Status (6)
Country | Link |
---|---|
US (1) | US8565494B2 (ja) |
EP (1) | EP2444933B1 (ja) |
JP (1) | JP5445584B2 (ja) |
KR (1) | KR101328358B1 (ja) |
CN (1) | CN102460507B (ja) |
WO (1) | WO2010146675A1 (ja) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012194770A (ja) * | 2011-03-16 | 2012-10-11 | Fujitsu Ltd | 生体認証装置、生体認証方法、および生体認証プログラム |
US20120307031A1 (en) * | 2011-05-30 | 2012-12-06 | Fujitsu Limited | Biometric information process device, biometric information process method, and computer readable medium |
WO2013161077A1 (ja) * | 2012-04-27 | 2013-10-31 | 富士通フロンテック株式会社 | 生体認証装置、生体認証プログラム及び生体認証方法 |
JP2014215868A (ja) * | 2013-04-26 | 2014-11-17 | 富士通株式会社 | 生体認証装置、生体認証プログラム、生体認証方法 |
CN104221052A (zh) * | 2012-03-28 | 2014-12-17 | 富士通株式会社 | 生物体认证装置、生物体认证方法、以及生物体认证程序 |
JP2015170300A (ja) * | 2014-03-10 | 2015-09-28 | 富士通株式会社 | 生体認証装置および携帯型電子装置 |
JP2023110379A (ja) * | 2022-01-28 | 2023-08-09 | 大日本印刷株式会社 | 電子情報記憶媒体、icカード、処理方法、及びプログラム |
Families Citing this family (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012014304A1 (ja) * | 2010-07-29 | 2012-02-02 | 富士通株式会社 | 生体認証装置および生体認証プログラム |
JP5846291B2 (ja) * | 2012-03-28 | 2016-01-20 | 富士通株式会社 | 生体認証装置、生体認証方法、および生体認証プログラム |
TW201342253A (zh) * | 2012-04-13 | 2013-10-16 | Hon Hai Prec Ind Co Ltd | 用於電子裝置中的指紋驗證方法及系統 |
US9348987B2 (en) * | 2012-04-19 | 2016-05-24 | Apple Inc. | Electronic device including finger-operated input device based biometric matching and related methods |
JP6089610B2 (ja) * | 2012-11-13 | 2017-03-08 | 富士通株式会社 | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム |
CH707230B1 (de) * | 2012-11-20 | 2016-02-29 | Frank Türen Ag | Türsystem mit berührungsloser Zutrittskontrolle und berührungsloser Türbedienung. |
JP6072890B2 (ja) * | 2013-02-15 | 2017-02-01 | Necソリューションイノベータ株式会社 | 類似判断の候補配列情報の選択装置、選択方法、およびそれらの用途 |
GB2514132A (en) * | 2013-05-14 | 2014-11-19 | Ibm | Two-directional biometric matching |
CN103997504B (zh) * | 2014-06-13 | 2017-11-10 | 谭知微 | 身份验证系统及身份验证方法 |
US9558392B2 (en) * | 2015-02-12 | 2017-01-31 | Korecen Co., Ltd. | Finger vein authentication system |
KR20170025083A (ko) * | 2015-08-27 | 2017-03-08 | 삼성전기주식회사 | 지문 감지 장치 및 이를 포함하는 전자 장치 |
US9749317B2 (en) | 2015-08-28 | 2017-08-29 | At&T Intellectual Property I, L.P. | Nullifying biometrics |
CN106295300B (zh) * | 2016-08-15 | 2019-04-26 | 深圳先进技术研究院 | 一种穿戴式智能设备身份认证方法及装置 |
US10693644B2 (en) * | 2017-06-23 | 2020-06-23 | International Business Machines Corporation | Single-input multifactor authentication |
JP6751072B2 (ja) * | 2017-12-27 | 2020-09-02 | 株式会社日立製作所 | 生体認証システム |
JP2020021406A (ja) * | 2018-08-03 | 2020-02-06 | 株式会社東海理化電機製作所 | 生体情報認証装置 |
JP2020038491A (ja) * | 2018-09-04 | 2020-03-12 | 株式会社東海理化電機製作所 | 生体情報認証装置 |
CN109858363B (zh) * | 2018-12-28 | 2020-07-17 | 北京旷视科技有限公司 | 一种狗鼻纹特征点的检测方法、装置、系统及存储介质 |
MX2022006009A (es) * | 2019-11-26 | 2022-06-22 | Spectrum Brands Inc | Sistema y metodo de inscripcion de usuarios de una cerradura biometrica inalambrica. |
KR20220108167A (ko) * | 2020-02-06 | 2022-08-02 | 후지쯔 가부시끼가이샤 | 인증 방법, 정보 처리 장치 및 인증 프로그램 |
CN113553989B (zh) * | 2021-08-04 | 2022-10-28 | 河南华辰智控技术有限公司 | 超级计算应用于生物识别特征分析的一种方法 |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0778248A (ja) * | 1993-06-30 | 1995-03-20 | Toshiba Corp | 個人認証装置 |
JPH08115425A (ja) * | 1994-10-17 | 1996-05-07 | Chuo Spring Co Ltd | 指紋照合装置 |
JPH08161491A (ja) | 1993-05-12 | 1996-06-21 | Sharp Corp | 指紋入力方法 |
JP2002222424A (ja) * | 2001-01-29 | 2002-08-09 | Nec Corp | 指紋照合システム |
JP2003337949A (ja) | 2002-05-21 | 2003-11-28 | Matsushita Electric Ind Co Ltd | 指紋照合装置 |
JP2007249339A (ja) | 2006-03-14 | 2007-09-27 | Fujitsu Ltd | 生体認証方法及び生体認証装置 |
JP2008217355A (ja) * | 2007-03-02 | 2008-09-18 | Fujitsu Ltd | 生体認証装置、生体認証プログラムおよび複合型生体認証方法 |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6757408B2 (en) * | 2002-01-25 | 2004-06-29 | Robert C. Houvener | Quality assurance and training system for high volume mobile identity verification system and method |
US6950536B2 (en) * | 2002-01-25 | 2005-09-27 | Houvener Robert C | High volume mobile identity verification system and method using tiered biometric analysis |
JP2006120026A (ja) | 2004-10-25 | 2006-05-11 | Hitachi Eng Co Ltd | 静脈認証装置および静脈認証方法 |
JP4671027B2 (ja) * | 2005-05-02 | 2011-04-13 | ソニー株式会社 | 認証装置、認証方法及びプログラム |
JP4952026B2 (ja) | 2006-03-31 | 2012-06-13 | 株式会社日立製作所 | 生体情報認証装置および認証方法 |
JP4786483B2 (ja) | 2006-09-14 | 2011-10-05 | 富士通株式会社 | 生体認証装置の生体誘導制御方法及び生体認証装置 |
JP4956131B2 (ja) | 2006-10-06 | 2012-06-20 | シャープ株式会社 | 生体認証装置及び方法、並びに生体認証処理プログラム |
JP4403426B2 (ja) * | 2007-01-09 | 2010-01-27 | サイレックス・テクノロジー株式会社 | 生体認証装置及び生体認証プログラム |
-
2009
- 2009-06-17 KR KR1020117029711A patent/KR101328358B1/ko active IP Right Grant
- 2009-06-17 WO PCT/JP2009/061037 patent/WO2010146675A1/ja active Application Filing
- 2009-06-17 JP JP2011519354A patent/JP5445584B2/ja not_active Expired - Fee Related
- 2009-06-17 EP EP09846168.4A patent/EP2444933B1/en active Active
- 2009-06-17 CN CN200980159911.6A patent/CN102460507B/zh active Active
-
2011
- 2011-12-14 US US13/326,035 patent/US8565494B2/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08161491A (ja) | 1993-05-12 | 1996-06-21 | Sharp Corp | 指紋入力方法 |
JPH0778248A (ja) * | 1993-06-30 | 1995-03-20 | Toshiba Corp | 個人認証装置 |
JPH08115425A (ja) * | 1994-10-17 | 1996-05-07 | Chuo Spring Co Ltd | 指紋照合装置 |
JP2002222424A (ja) * | 2001-01-29 | 2002-08-09 | Nec Corp | 指紋照合システム |
JP2003337949A (ja) | 2002-05-21 | 2003-11-28 | Matsushita Electric Ind Co Ltd | 指紋照合装置 |
JP2007249339A (ja) | 2006-03-14 | 2007-09-27 | Fujitsu Ltd | 生体認証方法及び生体認証装置 |
JP2008217355A (ja) * | 2007-03-02 | 2008-09-18 | Fujitsu Ltd | 生体認証装置、生体認証プログラムおよび複合型生体認証方法 |
Non-Patent Citations (1)
Title |
---|
See also references of EP2444933A4 |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012194770A (ja) * | 2011-03-16 | 2012-10-11 | Fujitsu Ltd | 生体認証装置、生体認証方法、および生体認証プログラム |
US20120307031A1 (en) * | 2011-05-30 | 2012-12-06 | Fujitsu Limited | Biometric information process device, biometric information process method, and computer readable medium |
US9412014B2 (en) * | 2011-05-30 | 2016-08-09 | Fujitsu Limited | Biometric information process device, biometric information process method, and computer readable medium |
CN104221052A (zh) * | 2012-03-28 | 2014-12-17 | 富士通株式会社 | 生物体认证装置、生物体认证方法、以及生物体认证程序 |
WO2013161077A1 (ja) * | 2012-04-27 | 2013-10-31 | 富士通フロンテック株式会社 | 生体認証装置、生体認証プログラム及び生体認証方法 |
JP2014215868A (ja) * | 2013-04-26 | 2014-11-17 | 富士通株式会社 | 生体認証装置、生体認証プログラム、生体認証方法 |
JP2015170300A (ja) * | 2014-03-10 | 2015-09-28 | 富士通株式会社 | 生体認証装置および携帯型電子装置 |
JP2023110379A (ja) * | 2022-01-28 | 2023-08-09 | 大日本印刷株式会社 | 電子情報記憶媒体、icカード、処理方法、及びプログラム |
JP7439843B2 (ja) | 2022-01-28 | 2024-02-28 | 大日本印刷株式会社 | 電子情報記憶媒体、icカード、処理方法、及びプログラム |
Also Published As
Publication number | Publication date |
---|---|
US20120082348A1 (en) | 2012-04-05 |
EP2444933A1 (en) | 2012-04-25 |
CN102460507B (zh) | 2014-12-03 |
CN102460507A (zh) | 2012-05-16 |
US8565494B2 (en) | 2013-10-22 |
KR20120023083A (ko) | 2012-03-12 |
KR101328358B1 (ko) | 2013-11-11 |
JPWO2010146675A1 (ja) | 2012-11-29 |
JP5445584B2 (ja) | 2014-03-19 |
EP2444933B1 (en) | 2019-07-10 |
EP2444933A4 (en) | 2017-04-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5445584B2 (ja) | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム | |
JP5228872B2 (ja) | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラムならびにコンピュータシステム | |
JP5477385B2 (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 | |
JP5799586B2 (ja) | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム | |
JP5196010B2 (ja) | 生体情報登録装置、生体情報登録方法及び生体情報登録用コンピュータプログラムならびに生体認証装置、生体認証方法及び生体認証用コンピュータプログラム | |
JP5699845B2 (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 | |
KR101930115B1 (ko) | 생체 인증 장치, 생체 인증 방법 및 생체 인증용 컴퓨터 프로그램 | |
JP5799960B2 (ja) | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム | |
KR101603469B1 (ko) | 생체 인증 장치, 생체 인증 방법 및 생체 인증용 컴퓨터 프로그램 | |
JP5812109B2 (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 | |
JP6467852B2 (ja) | 生体情報補正装置、生体情報補正方法及び生体情報補正用コンピュータプログラム | |
JP2008009753A (ja) | 生体認証方法及び生体認証システム | |
JP6024141B2 (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 | |
JP5915336B2 (ja) | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム | |
JP2015121874A (ja) | 生体認証装置、照合用データ検証方法及び照合用データ検証用コンピュータプログラム | |
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 | |
JP6056398B2 (ja) | 生体認証装置、生体認証方法及び生体認証用コンピュータプログラム | |
JP6079440B2 (ja) | 認証装置、認証方法および認証プログラム | |
US10528805B2 (en) | Biometric authentication apparatus, biometric authentication method, and computer-readable storage medium | |
JP2005275605A (ja) | 個人認証装置および個人認証方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200980159911.6 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09846168 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2011519354 Country of ref document: JP |
|
ENP | Entry into the national phase |
Ref document number: 20117029711 Country of ref document: KR Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2009846168 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |