Classifications

• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
  • G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
  • G06F21/31—User authentication
  • G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints

Definitions

• the present invention relates to a method and a biometric identification system.
• biometric identification are based on the comparison of biometric data of a candidate for identification and biometric data of users who have been previously stored for example in a database.
• the identification is validated when the biometric data of the candidate correspond to biometric data of one of the users whose biometric data are stored in the database.
• the biometric data results from the processing of biometric characteristics detected on the candidate and the users.
• the biometric characteristics concern, for example, fingerprints, iris patterns of the eyes, facial features ... and are generally captured by optical means connected to a computer processing unit.
• the optical means provide the computer processing unit with images in the form of signals, an image processing program of which extracts the biometric characteristics and codes them informally to form the biometric data.
• the computer processing unit hosts the database containing the identification data, including biometric data, of users who have been captured and stored beforehand, and executes a comparison program (or "matching") which compares the biometric data of the candidate stored biometric data to validate or refuse identification.
• Such identification systems are used, for example, at the entrance of places with restricted access.
• An identification system with one or more biometric capture devices each associated with a gate is disposed at the entrance of the restricted area to control the opening of the gates in case of validated identification.
• a disadvantage of such biometric identification systems lies mainly in the time taken by the comparison of biometric data-s, the comparison is even longer than the number of biometric data in the database is high.
• Another solution would be to divide the database into two groups between which the characteristics of the users are distributed. Each user would be informed of the group to which they were assigned and should indicate this at the time of capture using a capture device keyboard.
• This solution has the advantage of having a limited impact on the material cost. However, it assumes that the user remembers the group to which he has been assigned and still imposes the presence of a keyboard whose integration on the biometric capture device increases the cost of the latter.
• An object of the invention is to allow a reduction in the time required to identify a portion of the users of an identification system.
• a method of identifying persons from at least one database which is hosted on a computer unit and which groups together user identification data including data.
• biometric data relating to biometric characteristics of the users, the method comprising the steps of capturing and computer coding, at a predetermined identity check location, the biometric characteristics of an identification candidate to form biometric data of the candidate; comparing by performing a comparison program the biometric data of the candidate to the biometric data of the database and validating or rejecting the identification according to the result of the comparison.
• the data of the database is distributed in at least one first group of data and a second group of data, the comparison is made by first using the biometric data of the first group and then, if the comparison fails, using the biometric data of the second group.
• the method comprises a second step of capture and computer coding of biometric data in case of the comparison using the data of the first group, the computer coding performed during the second capture and coding step being more accurate than the computer coding performed in the first capture and coding step, the comparison using the data of the second group being made on the basis of the biometric data captured and encoded during the second capture and coding step.
• a relatively fast capture and coding step is used for regular users and a more accurate capture and coding step is used for non-regular users.
• the second group includes all the data of the database, that is to say also the data of the first group;
• the second capture and coding step is followed by a comparison of the data of the first group and the data from the second capture and coding step, and the comparison using the data of the second group being carried out in case of failure. this comparison of the data of the first group and the data from the second capture step.
• the method of the invention subjects the users of the first group having failed identification to the second capture step. If the first step of capture and coding is faster than the second and is normally sufficient in condition normal use to validate the identification, the second step of capture and coding, more accurate but usually slower, avoids a refusal of identification of a regular user that would result from a problem of capture or coding.
• the comparison using the data of the first group is carried out by means of a comparatively summary comparison method and the comparison using the data of the second group is carried out by means of a comparatively thorough comparison method.
• the comparison made on the data of the second group is slower but more reliable than that performed on the data of the first group, considering that the risk of fraud is greater vis-à-vis non-regular users.
• the distribution of data between the groups is modified according to a number of validated identifications stored for each user, the first group comprising the data of the users comprising a number of validated identifications greater than a predetermined threshold over a predetermined period .
• regular users will automatically be assigned to the first group and irregular users to the second group.
• the identification of regular users is validated more quickly than those of non-regular users. This results in an overall decrease in the time required for identification.
• the method comprises the step of assigning to at least one user a number of validated identifications greater than the number of identifications actually validated by said user. This makes it possible to define priority users by making them regular users even if they are not in reality.
• the distribution is specific to each site and takes into account the number of validated identification on each of the sites.
• Regular users of one site are not necessarily the same as those on the other site.
• each site thus has a group of its own regular users.
• the method comprises the step of periodically reducing the number of validated identifications for all users and / or the threshold is recalculated periodically.
• the invention also relates to a biometric identification system, comprising a processing computer unit connected to at least one biometric capture and coding unit for forming biometric data, the computer unit hosting a database that groups together user identification data including biometric data relating to biometric characteristics of the users and being arranged to execute at least one computer program for implementing the method of the above type.
• FIG. 1 is a schematic view of an identification system according to the invention.
• FIG. 2 is a block diagram showing the progress of the identification method according to the invention.
• biometric identification system according to the invention are here described in application to an identification by means of fingerprints. It goes without saying that other biometric characteristics are usable and in particular palm footprints, eye iris patterns, facial features, the shape of earplugs ...
• the system according to the invention is set up on two sites S1, S2, for example two sites of a company to which the members of the company must be able to access provided they are identified by the system. biometric identification.
• the identification system comprises a processing computer unit 1 connected to at least one unit 2 for capturing and encoding biometric characteristics.
• the capture and coding unit 2 comprises, in a manner known per se, an image sensor and an illumination member which are oriented towards a capture zone in which the user must present a finger to perform the capture. .
• the capture and coding unit 2 is associated with a gate 3.
• the computer unit 1 known in itself, has a memory hosting a database 10 which groups user identification data including biometric data relating to the biometric characteristics of the users.
• the memory contains at least one computer program for implementing the method according to the invention.
• Computer Unit 1 is arranged to execute this program.
• the computer program comprises a biometric data matching module (or “matching") known in itself and a control module of the lock 4 of the gate 3 according to the result of the identification.
• the capture and coding unit 2 is arranged in a manner known per se to detect biometric characteristics on the finger in question, encode them and transmit them to the processing computer unit 1 to form biometric data.
• the database 10 groups together identification data of the members of the company (the users) including biometric data relating to the biometric characteristics of the users.
• biometric data relating to the biometric characteristics of the users.
• one or more fingerprints of each user have been captured, coded and processed to form biometric data which are stored in the database with the names of the users, possibly rights of the user. access, an address, their functions, a telephone number, a name of manager ...
• the identification data of the users are assigned to two groups in the database 10.
• a first of these groups, Gl is intended to group the identification data of the regular users of the site SI, S2 concerned and the second of these Groups, G2, is intended to group the credentials of all users.
• the assignment to the groups is for example obtained by filling a field in the database.
• the assignment of the users to the first group is done a priori based on where the users are supposed to work and / or on the basis of their function.
• the database 10 further comprises a field intended to contain a number of validated identifications. This number is automatically incremented each validated identification.
• the identification method implemented during the execution of the program by the processing computer unit 2 begins with step 100 of capturing and computer code the characteristics of the fingerprint identification candidate to form biometric data of the candidate.
• a first comparison step 200 is then performed in which the candidate's biometric data is compared to the data of the first group G1.
• step 300 The result of the comparison is then appreciated (step 300).
• the biometric data of the candidate correspond to those of a user of the first group
• it is proceeded to the opening of the gate (step 400).
• a second capture and computer coding step 500 is then performed.
• the computer coding performed during the second capture and coding step 500 is more accurate than the computer coding performed during the first capture and coding step 100.
• the second capture and coding step 500 is followed by a second comparison step 600 during which the biometric data of the first group and the biometric data from the second capture and coding step 500 are compared.
• the result of the second comparison step 600 is appreciated (step 700):
• the gate is opened (step 400),
• a third comparative step 800 is performed in which the biometric data from the second capture and coding step 500 are compared with the biometric data of the second group.
• step 900 The result of the third comparison is appreciated (step 900):
• the gate is opened (step 400),
• the comparison steps 600 and 800 are performed using a more accurate comparison algorithm than for the first comparison step 200.
• the processing computer unit 2 controls the opening of the gate 3 and also increments the number of validated identifications stored in the database for the identified user (step 1000).
• the number of validated identifications memorized for each user is compared with a predetermined threshold.
• the predetermined threshold When the number of validated identifications of a user becomes greater than the predetermined threshold, the user is assigned to the first group. It is understood that the users of the first group are not necessarily the same in the two sites SI, S2.
• the threshold is for example calculated so as to minimize the following function F:
• the distribution of the data between the groups is thus modified according to the number of validated identifications memorized for each user, the first group comprising the data of the users comprising a number of validated identifications greater than the predetermined threshold over a predetermined period.
• the method further includes the step of periodically decreasing the number of validated identifications for all users.
• the method may further comprise the step of assigning to at least one user a number of validated identifications greater than the number of identifications actually validated by said user.
• This user can be for example the holder of a transversal function leading him to frequent both sites, for example a leader of the company.
• the computer system may have an architecture distinct from that described.
• the system may for example include a database hosted on a common server to which the computer units of each site are connected.
• the database includes a field with the number of validated identifications on each site and the group to which each user belongs for each site.
• the method may comprise only one capture and coding step.
• the comparison is then carried out in two phases:
• the comparison is made by comparing the biometric data of the candidate (resulting from the single capture and coding step) and the data of the first group G1;
• the comparison is made by comparing the biometric data of the candidate (resulting from the single capture and coding step) and the data of the second group G2.
• the comparison steps can be performed using the same algorithm.
• the second group may not understand the identification data of the first group.
• the distribution of the biometric data within the different groups is carried out with the use of individual statistics to each person. These individual statistics are incremented at each positive identification within the equipment. Frequent users of the device are thus favored to the detriment of those identifying themselves occasionally.

Landscapes

• Engineering & Computer Science (AREA)
• Computer Security & Cryptography (AREA)
• Theoretical Computer Science (AREA)
• Computer Hardware Design (AREA)
• Software Systems (AREA)
• Physics & Mathematics (AREA)
• General Engineering & Computer Science (AREA)
• General Physics & Mathematics (AREA)
• Collating Specific Patterns (AREA)

Applications Claiming Priority (2)

Publications (1)

Family

ID=49578426

Family Applications (1)

Country Status (4)

Families Citing this family (3)

Citations (2)

Family Cites Families (5)

• 2013
  • 2013-07-26 FR FR1357418A patent/FR3009106B1/fr active Active
• 2014
  • 2014-07-24 WO PCT/EP2014/065886 patent/WO2015011219A1/fr active Application Filing
  • 2014-07-24 EP EP14742226.5A patent/EP3025269A1/de not_active Withdrawn
  • 2014-07-24 US US14/906,445 patent/US10146920B2/en active Active

Patent Citations (2)

Non-Patent Citations (2)

Also Published As

Similar Documents

Legal Events