WO2006077764A1 - 閾値決定装置、方法及びプログラム並びに本人認証システム - Google Patents
閾値決定装置、方法及びプログラム並びに本人認証システム Download PDFInfo
- Publication number
- WO2006077764A1 WO2006077764A1 PCT/JP2006/300284 JP2006300284W WO2006077764A1 WO 2006077764 A1 WO2006077764 A1 WO 2006077764A1 JP 2006300284 W JP2006300284 W JP 2006300284W WO 2006077764 A1 WO2006077764 A1 WO 2006077764A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- threshold
- data
- individual
- distribution
- individual threshold
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1365—Matching; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
Definitions
- Threshold determination apparatus for Threshold determination apparatus, method and program, and personal identification system
- the present invention relates to, for example, a threshold value determining device that determines a threshold value of a matching evaluation value such as similarity for data such as fingerprints.
- Biometrics authentication using physical characteristics of an individual is known as a method of authenticating an individual. This is to check whether the physical features unique to the individual such as fingerprints and faces are the same as those registered, and if they are the same, authenticate as the person. For example, in the case of fingerprint authentication, an authorized! /, A person registers fingerprint data (template) for comparison in the authentication system in advance, and when it is desired to receive authentication, the user compares the input fingerprint data with the template. It is possible to determine whether the registrant is another person or not from the magnitude relationship between the matching evaluation value such as the degree of similarity and the predetermined threshold value.
- the threshold value is set to a strict value
- the false non-matching rate (the rate of occurrence of an error when it is determined that the same item is different) becomes large, and the convenience of the legitimate registrant is lost.
- the threshold value is set to a loose value
- the risk of non-registrants being authenticated increases as the false match rate (the rate at which an error occurs in which different items are judged to be the same) increases.
- the false match rate and the false non-match rate are determined by setting a threshold, and are in a trade-off relationship.
- Non-Patent Document 1 the relationship between the threshold and the false match rate is obtained by actually collating a large amount of test data and checking the false match rate when judged with various threshold values. It can be asked. If the threshold value is set so as to satisfy the false match rate required for the authentication system, the probability that the authentication system erroneously accepts others may be made equal to or less than the required value. it can.
- the threshold set in this way is set such that the average false match rate of all data is equal to or less than the required value, the false match rate for individual data differs depending on the data. Therefore, it is necessary to determine the threshold so that the false match rate for each data can be evaluated.
- the relationship between the threshold value and the false match rate or the false non match rate is determined for each individual, and the false match rate or the non match rate differs for each individual by determining the threshold. Solve the problem you end up with! /.
- Non-Patent Document 1 Japan Industrial Standards Committee (JISC), TSZTR No. “X0053”, TSZTR Name “Method for evaluating accuracy of fingerprint authentication system”
- Patent Document 1 Japanese Patent Application Laid-Open No. 2001-21309
- the first problem is that when the matching accuracy is evaluated as in Non-Patent Document 1, the accuracy of each individual data is not clear. The reason is that the accuracy evaluated by the method described in Non-Patent Document 1 is the average accuracy of all the data, so the false match rate in the case of determining the match Z non-match with the same threshold is different for each data. It is for.
- a second problem is that the method of Patent Document 1 can not determine the threshold at the time of system design.
- it is necessary to determine a threshold value by performing a large number of collations for each registration process in which the user registers template data in the authentication system.
- the reason is that the threshold value is also required for individual matching test. Therefore, it takes time and effort to enter a large number of data necessary for collation in registration processing which can not determine the threshold value in advance, and it takes time for registration processing by the collation processing of a large number of data. There's a problem.
- an object of the present invention is to provide a threshold value determination device and the like capable of defining a threshold value that can guarantee that the false match rate of each registered data is less than a predetermined value at a predetermined rate.
- Another object of the present invention is to provide a threshold determination device and the like capable of determining a threshold at design time.
- the threshold value determination device obtains a matching evaluation value among a plurality of data for each data, and the threshold value of the matching evaluation value satisfying a predetermined false match rate is an individual threshold.
- Individual threshold evaluation means for obtaining data for each data, and individual threshold distribution evaluation means for obtaining an individual threshold distribution of the data of individual data for each individual threshold obtained for the individual threshold evaluation means It is a feature of the present invention to have computing means for obtaining an overall threshold based on the individual threshold distribution obtained by the individual threshold distribution evaluation means.
- the threshold value thresholding device is provided with the individual threshold value evaluation means, the individual threshold value distribution evaluation means, and the threshold value determination means, and can achieve the predetermined false match rate at a predetermined rate.
- the first object of the present invention can be achieved by adopting such a configuration and defining a threshold value so that the individual nonconforming rate can achieve a predetermined value, not the average of the entire data.
- the second object of the present invention can be achieved by setting the threshold value using test data at the time of system design.
- the calculating means has a function of obtaining a threshold common to all of the plurality of data as a whole threshold based on the individual threshold distribution obtained by the individual threshold distribution evaluating means, or the individual threshold distribution evaluating means
- the system is configured to have one of the functions of determining the relationship between the overall threshold and the ratio of data having the individual threshold satisfying the overall threshold to all data based on the individual threshold distribution determined in Just do it.
- the matching evaluation value may be a similarity or a distance
- the data may be biological information identifying the user
- the biological information may be fingerprint data
- the person authentication system authenticates the person using the entire threshold obtained by the threshold determination device according to the present invention, or a relation obtained by the threshold determination device according to the present invention. Based on the above, authentication of the person is performed using the overall threshold value corresponding to the level of security required in the authentication.
- a matching evaluation value for another data is obtained for each data among a plurality of data, and the threshold value of the matching evaluation value satisfying a predetermined false match rate is used as an individual threshold value.
- the individual threshold evaluation step obtained for each data and the individual threshold for each data obtained in the individual threshold evaluation step the number of the data for each individual threshold? It is characterized in that an individual threshold distribution evaluation step of obtaining an individual threshold distribution and an operation step of calculating an overall threshold based on the individual threshold distribution obtained in the individual threshold distribution evaluation step are executed.
- a threshold value determination program determines, for each data, a comparison evaluation value among a plurality of data in a computer constituting a threshold value determination device for determining a threshold value, and the comparison satisfying a predetermined false match rate.
- the present invention can also be configured as follows. That is, (1) the similarity is evaluated for one of the test data and a plurality of data different from the data, and the ratio of the similarity exceeding the threshold does not exceed the required false match rate.
- the threshold may be determined for each data, and the appearance distribution of the threshold of the individual data may be evaluated to determine the relationship between the overall threshold and the rate at which the overall threshold exceeds the threshold of the individual data.
- the overall threshold may be determined. Effect of the invention
- the first effect is that it becomes possible to guarantee at a predetermined rate that a predetermined false match rate can be achieved for individual data.
- the reason is that the relationship between the threshold and the false match rate is not determined as the average relationship of all the data, but rather the threshold that can achieve the predetermined non-match rate for each data is determined, and the occurrence distribution of the threshold is found. This is because the overall threshold is determined to include percentage data.
- a second effect is that the threshold can be determined at design time.
- the reason is that the threshold value is determined from the distribution of threshold values that can achieve the predetermined threshold value in the test data, and information for each individual is not required at the time of registration processing.
- FIG. 1 is a block diagram showing a first embodiment of a threshold value determination apparatus according to the present invention. The following description will be made based on this drawing.
- the threshold determination device 10 of the present embodiment basically includes an individual threshold evaluation unit 11, an individual threshold distribution evaluation unit 12 and an overall threshold determination unit 13.
- the individual threshold evaluation means 11 calculates the similarity to other data among the plurality of data for each data, and calculates the threshold of the similarity satisfying the predetermined false match rate as the individual threshold for each data.
- the individual threshold distribution evaluation means 12 calculates an individual threshold distribution which is the number power of the data for each individual threshold, for the individual threshold of each data determined by the individual threshold evaluation means 11. Based on the individual threshold distribution determined by the individual threshold distribution evaluating unit 12, the whole threshold determining unit 13 determines a threshold common to all of the plurality of data as the whole threshold.
- the overall threshold value determining means 13 is used as computing means for determining the overall threshold value based on the individual threshold value distribution found by the individual threshold value distribution assessment means.
- the threshold determination device 10 is implemented in a computer 17 operating under program control.
- An output device 18 such as a display is connected to the computer 17.
- the computer 17 includes a test data storage unit 14, a request false match rate input unit 15 and a request guarantee ratio input unit 16 in addition to the individual threshold evaluation unit 11, the individual threshold distribution evaluation unit 12 and the overall threshold determination unit 13. .
- Each of these means operates as follows.
- the individual threshold value evaluation means 11 collates the test data stored in the test data storage means 14 and outputs a threshold value which can achieve the required false match rate input by the request false match rate input means 15. Evaluate each time.
- the individual threshold distribution evaluation unit 12 evaluates the appearance distribution for each individual threshold that can achieve the required false match rate obtained by the individual threshold evaluation unit 11. Based on the distribution of the individual thresholds obtained by the individual threshold distribution evaluation means 12, the whole threshold determination means 13 makes it possible to achieve the required false match rate at the rate input by the required guarantee rate input means 16. decide.
- the threshold determined by the overall threshold determination means 13 is output from the output device 18.
- FIG. 2 is a flowchart showing the operation of the threshold determination device 10.
- the test data storage means 14 stores test data (step A1), and the request false match rate input means 15 inputs the required false match rate (step A2).
- the order of step A1 and step A2 may be first.
- the individual threshold value evaluation means 11 obtains, for each data, a threshold value which is less than the false match rate input by the request false match rate input means 15 when collated with many other test data (step A3) ).
- the individual threshold distribution evaluation unit 12 examines the appearance distribution of the individual threshold of each data obtained by the individual threshold evaluation unit 11 (step A4).
- the required guarantee rate input unit 16 inputs a required value of a rate that can guarantee that the false match rate for each data is less than the required false match rate (step A5).
- Step A5 may be in the middle of steps A1 to A4 or any time before step A1.
- the overall threshold value determining means 13 determines the threshold value that can be guaranteed by the required guarantee rate with the required false match rate based on the distribution of the individual threshold values obtained by the individual threshold value distribution evaluation means 12 (step A6).
- the output device 18 outputs the determined threshold (step A7).
- the provision of the individual threshold value evaluation means 11, the individual threshold value distribution evaluation means 12 and the threshold value determination means 13 achieves a predetermined false match rate at a predetermined rate.
- a threshold can be defined. That is, the first object of the present invention can be achieved because the threshold value is set so that the data individual non-match rate which is not the average of all the data can achieve a predetermined value. Furthermore, the second object of the present invention can be achieved by setting the threshold using test data at the time of system design.
- FIG. 3 is a block diagram showing a second embodiment of the threshold value determination device according to the present invention. The following description will be made based on this drawing. However, the same parts as those in FIG.
- the threshold determination device 20 of the present embodiment basically includes an individual threshold evaluation unit 11, an individual threshold distribution evaluation unit 12, and a threshold guarantee ratio relation determination unit 21. Based on the individual threshold value distribution determined by the individual threshold value distribution evaluation means 12, the threshold value guarantee rate relationship determining means 21 determines the overall threshold value and the ratio of the data having the individual threshold value satisfying the overall threshold value to all data. Seek a relationship. In this embodiment, as the calculation means for obtaining the overall threshold value based on the individual threshold value distribution obtained by the individual threshold value distribution evaluation means, the threshold value maintenance is performed. It uses the certificate relationship determining means 21.
- the threshold determination device 20 is implemented in a computer 17 operating under program control.
- An output device 18 such as a display is connected to the computer 17.
- the computer 17 includes a test data storage unit 14 and a request false match rate input unit 15 in addition to the individual threshold value evaluation unit 11, the individual threshold distribution evaluation unit 12 and the threshold value guarantee rate relationship determination unit 21.
- Each of these means operates roughly as follows.
- the individual threshold evaluation unit 11, the individual threshold distribution evaluation unit 12, the test data storage unit 14 and the request false match rate input unit 15 operate in the same manner as the respective units in the first embodiment.
- the threshold guarantee ratio relation determination unit 21 can guarantee that the threshold and the request match ratio input by the request mismatch ratio input unit 15 are less than the threshold.
- the output device 18 outputs the relationship between the evaluation value obtained by the threshold guarantee ratio relationship determining means 21 and the guarantee ratio.
- FIG. 4 is a flowchart showing the operation of the threshold value determination device 20.
- the operation of the threshold value determination device 20 will be described based on FIG. 3 and FIG.
- each means in steps A1 to A4 is the same as the operation of each means in steps A1 to A4 in FIG.
- the threshold guarantee ratio relation determining means 16 is a ratio and threshold that can guarantee that the false match rate for each data is less than the required false match rate. Find the relationship of (step B5). Subsequently, the output device 18 outputs the relationship between the determined threshold value and the guarantee rate (step B6).
- the threshold value determination device 20 of the present embodiment in addition to the same effects as the first embodiment, authentication using the entire threshold value according to the required level can be realized.
- FIG. 5 is a block diagram showing an example of an identity authentication system using the threshold value determination device according to the present invention. The following description will be made based on this drawing.
- the personal identification system 30 uses a fingerprint as biometric information for identifying the individual, includes a collation device 31 and a determination unit 34, and uses the overall threshold value obtained by the threshold value determination device 10 of the first embodiment.
- Authentication of the The collation device 31 includes a fingerprint imaging unit 32 and a fingerprint collation unit 3 And 3.
- the fingerprint imaging unit 32 is, for example, a fingerprint scanner having a function of inputting a fingerprint electronically.
- the fingerprint collating unit 33 collates the fingerprint registered in advance with the newly input fingerprint using the provided collation algorithm.
- the determination unit 34 determines whether or not the person is the person using the overall threshold value input from the threshold value determination device 10 and the comparison result output from the fingerprint comparison unit 33.
- the personal identification system 30 is the same as the conventional one except that the global threshold obtained by the threshold determination device 10 is used, and thus the detailed description is omitted.
- the threshold value determination device 20 of the second embodiment is used, and based on the relationship obtained by the threshold value determination device 20, the height of the security required by the authentication The user may be authenticated using the entire threshold.
- the threshold determination devices 10 and 20 are provided outside the fingerprint collating unit 33 and the judging unit 34. This is merely an example.
- the threshold value determination devices 10 and 20 may be incorporated in the fingerprint comparison unit 33 or the determination unit 34.
- a personal computer is used as the computer 17.
- the memory in the personal computer is used as the test data storage means 14.
- values are entered from the keyboard connected to the personal computer.
- Fingerprint data is used as data, and the threshold used in the personal identification system that authenticates the person using fingerprint data is output.
- the requirement false match rate input means 15 inputs the required false match rate (request false match rate).
- the requirement false match rate refers to a fingerprint authentication apparatus that uses the threshold value determined in this embodiment for the probability of occurrence of an error (mismatch rate) in which different fingerprint data are erroneously determined to be the same fingerprint data.
- the predetermined value is required to be smaller than this value.
- a request false match rate input means 15 Enter 1 / 10,000.
- the request false match rate input means 15 can be stored in advance in a memory by inputting a predetermined value from the keyboard, or can be input through communication or an external recording medium.
- the test data storage unit 14 stores fingerprint data for a test.
- the test data storage means 14 can store not only fingerprint data for test in a memory in the personal computer, but also in a recording medium such as a hard disk or a DVD medium. Alternatively, it may be stored in a storage device external to the personal computer, communicated through a network, and sequentially input only data necessary for collation, so that only the necessary amount may be stored. it can.
- the required guarantee rate (required guarantee rate) is input.
- the required guarantee rate is the percentage of the total fingerprint data that, when each fingerprint data is compared with many fingerprints, the false match rate for each data is guaranteed to be smaller than the required false match rate. It is. For example, if the request false match rate is 1/1000, we want to guarantee that the false match rate is smaller than 10,000 for fingerprint data of 99% or more, that is, the false match rate is larger than the required value. If you want to limit the possibility to less than 1%, the required guarantee rate will be 99%.
- the required guarantee rate input means 16 can also be stored in advance in a memory by inputting a predetermined value from the keyboard, or can be input through communication or an external recording medium.
- the individual threshold value evaluation unit 11 obtains, for each fingerprint data of test data, a threshold value at which the false match rate becomes the required false match rate. That is, for each fingerprint data, the similarity to other test data is evaluated. Then, for each fingerprint data, a threshold is determined such that the probability that the degree of similarity exceeds the threshold is less than or equal to the required false match rate.
- the request false match rate is 1 / 10,000 and a certain fingerprint data (fingerprint data 1) is compared with 60000 pieces of fingerprint data.
- Matching with a fingerprint of 6 000 or less that is equal to or less than 1 / 10,000 is equal to or higher than a certain value, but with matching to the remaining 59994 fingerprints, the similarity is lower than that value
- a threshold for fingerprint data 1 is determined. For example, when fingerprint data 1 and 60000 pieces of fingerprint data are compared, the highest degree of similarity with the highest strength is 0.22, 0.20, 0.19, 0.18, 0.18, 0.17 when the six are arranged in order, and the remaining 59,994 Is less than 0.17.
- the threshold value for fingerprint data 1 may be larger than 0.17.
- the threshold value for each data may be set arbitrarily within this range, but if the threshold value is raised, the false non-matching rate (the rate at which the same finger is incorrectly determined to be a different finger) increases, and the convenience is impaired. Therefore, the smallest value within this range may be 0.17.
- the threshold for each data is different depending on the data. For example, when fingerprint data 2 is compared with 60000 pieces of fingerprint data, the most similar strengths are also obtained by arranging six pieces in order of 0.34, 0.32, 0.30, 0.29, 0.28, 0.25, and the remaining 59,994 Is less than 0.25, the individual threshold for fingerprint data 2 is in the range of 0.25 or more. Since the individual threshold differs depending on the data, as shown in Table 1 below, the threshold for that data is determined for each data.
- Fingerprint data 3 0.10 Fingerprint data N 0.23
- the threshold for each data will be referred to as an individual threshold.
- the frequency distribution of the individual threshold for each data determined by the individual threshold evaluation unit 11 is determined.
- the frequency distribution of individual thresholds is, for example, as shown in Fig. 6 [1], 1% in all data for which the individual threshold is smaller than 0.0, 14% for more than 0.01 and less than 0.1 for 14%, more than 0.1 for less than 0.1. 60%, 0.2% or more and less than 0.29 are 14%, and 0.29 or more is 1%, and the individual threshold value indicates how much of all the data is.
- the overall threshold determination means 13 determines the threshold with which the required false match rate can be guaranteed by the required guarantee rate from the distribution of the individual threshold obtained by the individual threshold distribution evaluation means 12. For example, the requirement false match rate is 1/1000, the requirement guarantee rate is 1%, and the distribution of individual thresholds is 1% for data with an individual threshold of 0.29 or more as shown in Fig. 6 [1]. In this case, the overall threshold may be 0.29 or more. In this case, the ratio of individual false match rate for each fingerprint data being less than 10,000 is as shown in Fig. 6 [2]. The false match rate for each data is the required false match rate for all data: 10,000 minutes What is greater than 1 will be less than 1%.
- the overall threshold determined by the overall threshold determination means 13 is output from the output device 18.
- output The device 18 can also display on a display device such as a display or can record and output it in a storage device.
- the false match rate for each data can be obtained by inputting the overall threshold value displayed on the display device or by the fingerprint authentication apparatus reading out the overall threshold value recorded in the storage device. It is possible to design a fingerprint authentication device that has a required guarantee rate or more and a required false match rate or less.
- the present example corresponds to the above-described second embodiment.
- the following description is based on FIG. 3, FIG. 4 and FIG.
- the present embodiment differs from the first embodiment in that it does not include the required guarantee ratio input unit 15 (FIG. 1) and that it includes the threshold guaranteed ratio relation determining unit 21.
- the threshold value guarantee rate relationship determining means 21 obtains the relation between the threshold value and the guarantee rate from the individual threshold value distribution obtained by the individual threshold value distribution evaluating means 12.
- the relationship between the threshold value and the guarantee rate is the relationship between how much the threshold value can guarantee that the false match rate is smaller than the required false match rate with a certainty.
- the individual threshold distribution is 4% when it is 0.25 or more and less than 0.29, 0.5% when it is 0.29 or more and less than 0.295, and 0.5% when it is 0.295 or more. If the individual threshold value is 0.25 or more, 5% (0.25 or less is 95%), 0.29 or more is 1% (0.29 or less is 99%), 0.295 or more is 0.5% (It is assumed that 99.5% were less than 0.295).
- setting the threshold to 0.25 can guarantee that the individual mismatch rate for each data is smaller than the required false match rate by 95%
- setting the threshold to 0.29 separates each data. It is possible to guarantee that the false match rate is 99% smaller than the required false match rate, and setting the threshold value to 0.295 guarantees that the individual false match rate for each data is smaller than the required false match rate by 99.5%. it can.
- the output device 18 outputs the relationship between the threshold and the guarantee rate. If the fingerprint authentication apparatus uses the relationship between the threshold and the guarantee rate, the guarantee rate can be changed and operated by switching the threshold according to the application for authentication. For example, when fingerprint authentication is performed in an in-house information system, a threshold that can be guaranteed to 99.9% is used for authentication for accessing highly confidential information such as personnel relations, so that the degree of secrecy is high. It can be switched according to the secrecy of the information to be accessed, such as using a threshold that can be guaranteed against 99% for authentication for access.
- the first and second embodiments have been described using fingerprint data as data.
- a method of matching fingerprints there is a method using feature points such as ridge end points and bifurcation points.
- feature points such as ridge end points and bifurcation points.
- the number of feature points differs for each fingerprint, there may be variations in the false match rate for each fingerprint data due to the number of feature points.
- a fingerprint authentication system can be constructed that can guarantee the false match rate with the required guarantee rate even if the number of feature points differs for each fingerprint. it can.
- biometric information data that can identify another person, such as force face data, iris data, and vein data described using fingerprint data as data.
- biometric information has a large variation among individuals. Therefore, there is a possibility that the variation in false match rate is large for each individual.
- biometric information authentication system capable of guaranteeing a false match rate at a required guarantee rate.
- biometric information such as fingerprint data but also other authentication data can be used.
- the similarity is used as an evaluation value of the result of comparing two fingerprint data, and the larger the similarity, the more similar the two fingerprint data are, and the larger the threshold, the larger the value. It was explained that the conditions would be severe. However, as the evaluation value, use a scale that indicates that the smaller the numerical value is the same, such as the distance between the two data, and the smaller the numerical value, the more severe the threshold U.
- the present invention determines a threshold that can guarantee the required false match rate with the required guarantee rate. It can be used for the threshold determination device.
- the threshold determined by this device can be used to design a personal identification system.
- FIG. 1 is a block diagram showing a first embodiment of a threshold value determination device according to the present invention.
- FIG. 2 A flowchart showing the operation of the threshold value determination device of FIG.
- FIG. 3 is a block diagram showing a second embodiment of the threshold value determination device according to the present invention.
- FIG. 4 is a flowchart showing the operation of the threshold value determination device of FIG. 3;
- FIG. 5 is a block diagram showing an example of a person authentication system using a threshold value determination device according to the present invention.
- FIG. 6 [1] is a graph showing an example of the appearance frequency distribution of the individual threshold value for each fingerprint data.
- FIG. 6 [2] is a graph showing an example of the ratio of fingerprint data for which the individual false match rate is equal to or higher than a predetermined value and fingerprint data for values smaller than the predetermined value when the overall threshold is determined.
- FIG. 6 [3] is a graph showing an example of the appearance frequency distribution of the individual threshold value for each fingerprint data.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Human Computer Interaction (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Probability & Statistics with Applications (AREA)
- Collating Specific Patterns (AREA)
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US11/792,840 US7991204B2 (en) | 2005-01-21 | 2006-01-12 | Threshold determining device, method and program, and personal authentication system |
JP2006553865A JP4899868B2 (ja) | 2005-01-21 | 2006-01-12 | 閾値決定装置、方法及びプログラム並びに本人認証システム |
EP06711607.9A EP1847959B1 (en) | 2005-01-21 | 2006-01-12 | Threshold determining device, method and program, and person identifying system |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2005013493 | 2005-01-21 | ||
JP2005-013493 | 2005-01-21 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2006077764A1 true WO2006077764A1 (ja) | 2006-07-27 |
Family
ID=36692154
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2006/300284 WO2006077764A1 (ja) | 2005-01-21 | 2006-01-12 | 閾値決定装置、方法及びプログラム並びに本人認証システム |
Country Status (5)
Country | Link |
---|---|
US (1) | US7991204B2 (ja) |
EP (1) | EP1847959B1 (ja) |
JP (1) | JP4899868B2 (ja) |
CN (1) | CN100524360C (ja) |
WO (1) | WO2006077764A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012017589A1 (ja) * | 2010-08-05 | 2012-02-09 | パナソニック株式会社 | 顔画像登録装置および方法 |
US9083058B2 (en) | 2008-02-29 | 2015-07-14 | Mitsubishi Chemical Corporation | Nonaqueous electrolytic solution and nonaqueous-electrolyte battery |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101527711B1 (ko) * | 2007-12-20 | 2015-06-11 | 코닌클리케 필립스 엔.브이. | 템플릿 보호 시스템들에서의 분류 임계치들의 규정 |
JP4962813B1 (ja) * | 2012-01-23 | 2012-06-27 | 日本テクノ株式会社 | アラーム出力装置、アラーム出力方法 |
CN104794448A (zh) * | 2013-01-23 | 2015-07-22 | 深圳市亚略特生物识别科技有限公司 | 指纹验证*** |
CN105404807B (zh) * | 2015-12-08 | 2019-02-05 | Oppo广东移动通信有限公司 | 提升指纹识别性能的方法、装置及移动终端 |
TWI806030B (zh) * | 2021-03-31 | 2023-06-21 | 瑞昱半導體股份有限公司 | 應用於人臉辨識系統的處理電路及處理方法 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06208611A (ja) * | 1993-01-11 | 1994-07-26 | Toshiba Corp | 個人認証装置 |
JPH06274628A (ja) * | 1993-03-18 | 1994-09-30 | Fujitsu Ltd | 本人照合推定方法 |
JP2000182057A (ja) * | 1998-12-18 | 2000-06-30 | Nec Corp | 個人識別方法、個人識別装置および記録媒体 |
JP2000215313A (ja) * | 1999-01-22 | 2000-08-04 | Mitsubishi Electric Corp | デ―タ識別方法およびその装置 |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001021309A (ja) | 1999-07-12 | 2001-01-26 | Toshiba Tec Corp | 個体認証方法及び個人認証方法 |
JP2003067340A (ja) * | 2001-08-28 | 2003-03-07 | Mitsubishi Electric Corp | 認証の選択システム、認証システム |
-
2006
- 2006-01-12 JP JP2006553865A patent/JP4899868B2/ja active Active
- 2006-01-12 WO PCT/JP2006/300284 patent/WO2006077764A1/ja active Application Filing
- 2006-01-12 EP EP06711607.9A patent/EP1847959B1/en active Active
- 2006-01-12 US US11/792,840 patent/US7991204B2/en active Active
- 2006-01-12 CN CNB2006800026838A patent/CN100524360C/zh active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06208611A (ja) * | 1993-01-11 | 1994-07-26 | Toshiba Corp | 個人認証装置 |
JPH06274628A (ja) * | 1993-03-18 | 1994-09-30 | Fujitsu Ltd | 本人照合推定方法 |
JP2000182057A (ja) * | 1998-12-18 | 2000-06-30 | Nec Corp | 個人識別方法、個人識別装置および記録媒体 |
JP2000215313A (ja) * | 1999-01-22 | 2000-08-04 | Mitsubishi Electric Corp | デ―タ識別方法およびその装置 |
Non-Patent Citations (1)
Title |
---|
See also references of EP1847959A4 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9083058B2 (en) | 2008-02-29 | 2015-07-14 | Mitsubishi Chemical Corporation | Nonaqueous electrolytic solution and nonaqueous-electrolyte battery |
WO2012017589A1 (ja) * | 2010-08-05 | 2012-02-09 | パナソニック株式会社 | 顔画像登録装置および方法 |
JP2012037995A (ja) * | 2010-08-05 | 2012-02-23 | Panasonic Corp | 顔画像登録装置および方法 |
US9092660B2 (en) | 2010-08-05 | 2015-07-28 | Panasonic Intellectual Property Management Co., Ltd. | Face image registration device and method |
Also Published As
Publication number | Publication date |
---|---|
US7991204B2 (en) | 2011-08-02 |
JP4899868B2 (ja) | 2012-03-21 |
JPWO2006077764A1 (ja) | 2008-08-07 |
CN100524360C (zh) | 2009-08-05 |
US20080089562A1 (en) | 2008-04-17 |
EP1847959B1 (en) | 2013-11-27 |
EP1847959A1 (en) | 2007-10-24 |
CN101107627A (zh) | 2008-01-16 |
EP1847959A4 (en) | 2011-10-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR100528328B1 (ko) | 사용자 인증 방법 및 장치 | |
Crawford | Keystroke dynamics: Characteristics and opportunities | |
US7356168B2 (en) | Biometric verification system and method utilizing a data classifier and fusion model | |
Ko | Multimodal biometric identification for large user population using fingerprint, face and iris recognition | |
US20080212846A1 (en) | Biometric authentication using biologic templates | |
Saevanee et al. | Text-based active authentication for mobile devices | |
US7492925B2 (en) | Biometric identification data protection | |
US20070183632A1 (en) | Methods of identifier determination and of biometric verification and associated systems | |
NL1037554C2 (en) | System and method for verifying the identity of an individual by employing biometric data features associated with the individual as well as a computer program product for performing said method. | |
WO2006077764A1 (ja) | 閾値決定装置、方法及びプログラム並びに本人認証システム | |
Ho | Tapdynamics: strengthening user authentication on mobile phones with keystroke dynamics | |
JP2013122679A (ja) | 安全性の高い生体認証システム | |
Schclar et al. | User authentication based on representative users | |
Dahel et al. | Accuracy performance analysis of multimodal biometrics | |
Dahiya et al. | Biometrics security concerns | |
US9122857B1 (en) | Authenticating a user in an authentication system | |
JP5903257B2 (ja) | 認証システム、登録装置、及び認証装置 | |
US20100208950A1 (en) | Biometric identification data protection | |
Nair et al. | An approach to improve the match-on-card fingerprint authentication system security | |
Saifan et al. | A Survey of behavioral authentication using keystroke dynamics: Touch screens and mobile devices | |
Kalra et al. | A survey on multimodal biometric | |
US20070233667A1 (en) | Method and apparatus for sample categorization | |
Jagadamba et al. | A secured authentication system using an effective keystroke dynamics | |
CN113132380A (zh) | 一种密码算法的安全性评估***、方法、设备及介质 | |
CN112232443A (zh) | 身份认证方法、装置、设备及存储介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 11792840 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2006553865 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2006711607 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 200680002683.8 Country of ref document: CN |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWP | Wipo information: published in national office |
Ref document number: 2006711607 Country of ref document: EP |