JP2003044857A - System and device for matching fingerprint - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently perform fingerprint matching with high matching performance by selecting a matching algorithm corresponding to various finger conditions or various fingerprint reading conditions in a fingerprint matching system for authenticating an individual with fingerprint image matching. SOLUTION: A fingerprint reader 15 reads the fingerprint image data of a user purchasing a portable telephone 20, also decides a pattern of the fingerprint and a characteristic of the dispersion of reading positions, etc., and selectively determines a fingerprint matching algorithm that fits this fingerprint characteristic. A fingerprint matching processing program fitting the fingerprint image is rewritten and registered together with the read finger print image data of the user purchasing the telephone in the portable telephone 20 by a writing device 17. In performing, for example, person himself/herself authentication processing based on a fingerprint at a connected site on a communication network by utilizing the portable telephone 20, fingerprint matching processing software registered in the portable telephone 20 performs fingerprint matching between the telephone user and the registered fingerprint.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fingerprint collation system and a fingerprint collation device for authenticating an individual by collating fingerprint images.

[0002]

2. Description of the Related Art Conventionally, as one method for authenticating an individual, a fingerprint image of an individual to be authenticated is read and collated with a registered fingerprint image, and the degree of similarity is judged and registered. 2. Description of the Related Art A fingerprint collation system that determines whether the fingerprint image is the original person or another person is considered.

As a kind of matching algorithm in such a fingerprint matching system, there are a feature point extraction matching method, an image matching method, a frequency analysis method and the like. Basically, one fingerprint matching system has one matching algorithm. Is used to verify the fingerprints of various human fingers to be authenticated.

[0004]

However, with respect to a human finger, there are significant individual differences such as size and shape, and the finger can be obtained by placing or pressing the finger on the reading window at the time of reading. Since fingerprint images are different, if one finger is used to match fingers under various conditions, there is a problem that the false rejection rate and the false acceptance rate may deteriorate depending on the conditions.

The present invention has been made in view of the above problems, and selects a matching algorithm according to various finger conditions or various fingerprint reading conditions to efficiently perform fingerprint matching with high matching performance. An object of the present invention is to provide a fingerprint collation system and a fingerprint collation device that can be performed.

[0006]

That is, a fingerprint collation system according to a first aspect of the present invention provides a fingerprint reading means for reading a fingerprint image of an electronic device user, and a fingerprint reading means for reading the fingerprint image by the fingerprint reading means. A characteristic determination unit that determines the characteristic of the electronic device user, a matching algorithm selection unit that selects a fingerprint matching algorithm according to the characteristic of the electronic device user determined by the characteristic determination unit, and a fingerprint reading unit. Registration means for registering, in the electronic device, a fingerprint matching processing program according to the fingerprint matching algorithm selected by the matching algorithm selecting means together with the fingerprint image data of the electronic device user.

In such a fingerprint collation system according to the first aspect of the present invention, when the fingerprint image of the electronic device user is read by the fingerprint reading means, the fingerprint pattern of the electronic device user or the amount of misalignment accompanying the reading. The characteristics such as the above are determined, and the fingerprint matching algorithm according to the characteristics of the electronic device user is selected. Then, a fingerprint collation processing program according to a fingerprint collation algorithm selected according to the user characteristic is registered in the electronic device together with the fingerprint image data of the electronic device user read by the fingerprint reading means, and registered in the electronic device. The fingerprint collation processing program is executed to collate the fingerprint between the device user and the registered fingerprint.

Further, in the fingerprint collation device according to the second aspect of the present invention, the fingerprint acquisition means for acquiring the fingerprint image of the person to be collated, and the characteristic of the person to be collated with the fingerprint image obtained by the fingerprint acquisition means. According to the characteristic determination means for determining, the fingerprint matching algorithm selected by the matching algorithm selecting means, the fingerprint matching algorithm selecting means for selecting the fingerprint matching algorithm according to the characteristics of the matching target person judged by the characteristic judging means. And a fingerprint collation unit for collating the fingerprint image data acquired by the fingerprint acquisition unit by using the fingerprint collation processing program.

In the fingerprint collating apparatus according to the second aspect of the present invention, when the fingerprint image of the collation target person is acquired by the fingerprint acquisition means, the fingerprint image of the collation target person is acquired along with the acquisition of the fingerprint image. The characteristics such as the print pattern or the amount of positional deviation due to reading are determined, and the fingerprint matching algorithm according to the characteristics of the matching target person is selected. Then, the collation process of the fingerprint image data acquired by the fingerprint acquisition unit is performed based on the fingerprint collation processing program according to the fingerprint collation algorithm selected according to the characteristic of the collation target person.

[0010]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings.

(First Embodiment) FIG. 1 is a view showing the arrangement of a fingerprint collation system according to the first embodiment of the present invention.

This fingerprint collation system is composed of a terminal device (sales terminal) 10 of a mobile phone store and a mobile phone 20 with a fingerprint collation function sold by this store.

The sales terminal 10 is composed of a computer and has a CPU (central process) as a central control unit.
ing unit) 11.

The CPU 11 controls the operation of each section by using the RAM 13 as a work memory in accordance with a sales terminal control program stored in advance in a storage device 12 including a hard disk and a flash memory.
In addition to the storage device 12, the RAM 13, the input / output device 14, the fingerprint reading device 15, the sensor unit 16,
The writing device 17 is connected.

The storage device 12 stores the sales terminal control program that controls the entire operation of the sales terminal 10, and reads the fingerprint of the user who purchased the mobile phone 20 as the mobile phone 20 is sold. A fingerprint registration program for determining a fingerprint matching algorithm suitable for the characteristics and writing and registering the fingerprint image data and matching processing software (program) in the mobile phone 20, and a plurality of types of fingerprint matching algorithms (feature point extraction matching method). / Image matching method / Frequency analysis method, etc., corresponding matching processing software (program) is stored.

The input / output device 14 comprises an input device such as a keyboard and a mouse and a display output device such as a liquid crystal display.

The fingerprint reader 15 is a CCD (charge coupled device) that reads a fingerprint of a human finger as an optical image.
e) A sensor is provided, and the fingerprint image data read by this CCD sensor is temporarily stored in the fingerprint image memory in the RAM 13.

A finger positioning mechanism for picking up a uniform and good fingerprint image in the fingerprint reading device 15 will be described later.

The sensor unit 16 is provided in the vicinity of the fingerprint reading surface of the fingerprint reading device 15, and various sensors for determining conditions such as skin quality and elasticity of the finger and conditions such as how the finger is placed and size. It is equipped with various sensors for determining whether, for example, the elasticity of the finger skin is measured by a tactile sensor, and the moisture content of the finger skin and the height of ridges are measured by a capacitance sensor. To be measured.

The conditions such as the type of fingerprint pattern of the finger, the size of the finger or the variation of the fingerprint reading position are determined based on the fingerprint image read by the CCD sensor of the fingerprint reading device 15.

The contents of each individual judgment regarding the nature of the finger and the contents of each individual judgment regarding how to place the finger will be described later.

The writing device 17 is communicatively connected to the mobile phone 20 by a short-distance wireless technology such as a dedicated cable or bluetooth, and the fingerprint image data of the telephone purchase user read by the fingerprint reading device 15 through the writing device 17. And fingerprint matching processing software (program) with matching algorithm suitable for the characteristics of the fingerprint image
Are both written and registered in the mobile phone 20.

In other words, in this fingerprint collation system, the fingerprint collation processing software (program) of the collation algorithm according to the fingerprint image data of the telephone purchasing user and the individual characteristics and reading characteristics of the fingerprint is purchased.
By being registered in, it becomes possible to more accurately and efficiently perform personal identification processing using fingerprints on the communication network using the mobile phone 20.

Next, the fingerprint registration function of the fingerprint collation system having the above configuration will be described.

FIG. 2 is a flow chart showing a fingerprint registration process by the fingerprint collation system.

With the sale of the mobile phone 20 with the fingerprint collation function, first, the fingerprint reading device 15 of the sales terminal 10 reads the fingerprint image of the user who purchased the phone (step S1).

Then, based on the fingerprint image data and various sensor signals in the sensor unit 16, the personal characteristics (fingerprint pattern etc.) and the reading characteristics (variation of the reading position) of the fingerprint are judged (step SA).

Details of this characteristic determination process (step SA) will be described later.

Then, a matching algorithm suitable for the fingerprint characteristic of the telephone purchasing user determined according to the characteristic determination processing is selected and determined (step S2).

This collation algorithm is selected by, for example, preliminarily sampled fingerprint image data having different fingerprint characteristics, performing fingerprint collation processing by a plurality of types of collation algorithms, and obtaining corresponding authentication rate data. For the fingerprint characteristic of, the matching algorithm with the highest authentication rate is performed based on the associated table.

In this way, when the matching algorithm suitable for the fingerprint characteristic of the telephone purchasing user is selected and determined, the fingerprint image data of the telephone purchasing user read in step S1 is read from the fingerprint image memory in the RAM 13. At the same time, fingerprint collation processing software (program) of the collation algorithm determined according to the fingerprint characteristic is read from the storage device 12, and the writing device 17
The data is transmitted to the mobile phone 20 via and is registered for writing (step S3).

Therefore, according to the fingerprint registration function of the fingerprint collation system having the above-described configuration, the fingerprint reading device 15 reads the fingerprint image data of the user who purchases the mobile phone 20, and the characteristics such as the fingerprint pattern and reading position variation. Is determined, and a fingerprint matching algorithm suitable for this fingerprint characteristic is selected and determined. Since the fingerprint collation processing software (program) suitable for the fingerprint image is written and registered in the mobile phone 20 by the writing device 17 together with the fingerprint image data read by the user who purchased the phone, the mobile phone 20 is used. For example, when performing the personal identification processing by the fingerprint at the connection site on the communication network, by performing the fingerprint matching between the telephone user and the registered fingerprint by the fingerprint matching processing software (program) registered in the mobile phone 20, A more accurate and efficient authentication process can be performed.

(Second Embodiment) FIG. 3 is a view showing the arrangement of a fingerprint collation apparatus according to the second embodiment of the present invention.

This fingerprint collation device is composed of a collation server 30 on the communication network.
0 authenticates the user by using the fingerprint of the user of the communication terminal 40 such as a mobile phone or a personal computer, which is accessed via the network, as a collation with the registered fingerprint.

The collation server 30 is constituted by a computer having a CPU (central processing unit) as its central control unit.

The collation server 30 includes a server control program that controls the overall operation of the collation server 30, a fingerprint authentication program that authenticates fingerprints acquired from the communication terminal 40, and a plurality of fingerprint collation algorithms (features). A program memory 31 in which matching processing software (program) corresponding to the point extraction matching method / image matching method / frequency analysis method) is stored, a plurality of communication terminals 40 accessible on the communication network ... Fingerprints of respective users The registration data memory 32 in which the image data is registered in association with each terminal ID together with the characteristic data of each fingerprint, and the fingerprint image data to be collated acquired from the communication terminal 40 is accumulated together with the characteristic data and stored. A data memory 33 is provided.

On the other hand, the communication terminal 40 is also composed of a computer having a CPU (central processing unit) as its central control unit.

The communication terminal 40 is provided with a key input device, a display device, and a storage device, and a CCD (charge coupled device) for reading a fingerprint of a human finger as an optical image.
e) A fingerprint reading device 41 equipped with a sensor, various sensors provided near the fingerprint reading surface of the fingerprint reading device 41 for determining conditions such as skin quality and elasticity of the finger, and how to place and size the finger. A sensor unit 42 provided with various sensors for judging conditions such as
Is provided.

In this sensor unit 42, for example, the elasticity of the finger skin is measured by a tactile sensor, and the water content of the finger skin and the height of the ridge are measured by a capacitance sensor.

The conditions such as the type of fingerprint pattern of the finger, the size of the finger, and the variation of the fingerprint reading position are determined based on the fingerprint image read by the CCD sensor in the fingerprint reading device 41.

The contents of each individual judgment regarding the nature of the finger and the contents of each individual judgment regarding how to place the finger will be described later.

A finger positioning mechanism for picking up a uniform and good fingerprint image in the fingerprint reading device 41 will be described later.

Next, the fingerprint collation function of the fingerprint collation device having the above configuration will be described.

FIG. 4 is a flow chart showing a fingerprint authentication process by the fingerprint collation device.

In the collation server 30, in order to authenticate the user of the communication terminal 40 connected to the communication network, the fingerprint image data to be collated read by the fingerprint reading device 41 of the communication terminal 40 and the sensor unit 42 are used. When various sensor signals relating to the acquired read fingerprint are received, the collated fingerprint image data and various sensor data relating to the fingerprint are accumulated in the accumulated data memory 33 (step B1).

Then, based on the collation fingerprint image data and various sensor data stored in the accumulated data memory 33, the individual characteristics (fingerprint pattern, etc.) and the reading characteristics (variation of the reading position, etc.) of the collation fingerprint are judged. (Step BA).

Details of this characteristic judgment processing (step BA) will be described later.

Then, the collation algorithm suitable for the fingerprint characteristic of the terminal user determined according to this characteristic determination processing is selected and determined (step B2).

This collation algorithm is selected by, for example, preliminarily sampled fingerprint image data having different fingerprint characteristics, performing fingerprint collation processing by a plurality of types of collation algorithms, and obtaining respective corresponding authentication rate data. For the fingerprint characteristic of, the matching algorithm with the highest authentication rate is performed based on the associated table.

In this way, when the collation algorithm suitable for the fingerprint characteristic of the terminal user is selected and determined, the fingerprint collation processing software (program) of the collation algorithm is read from the program memory 31, and this fingerprint collation processing software ( According to the program), the accumulated data memory 3
The collation fingerprint image data of the access terminal user stored in No. 3 is collated with the fingerprint image data associated with the corresponding terminal ID registered in the registration data memory 32 (step B3).

Further, the characteristic data regarding the collated fingerprint image stored in the accumulated data memory 33 is judged to be identical with the characteristic data regarding the fingerprint image associated with the corresponding terminal ID registered in the registration data memory 32. For example, it is determined whether or not there is an abnormality in characteristics other than the fingerprint pattern, such as different skin characteristics or different ridge characteristics of the fingerprint (step B4).

Then, it is verified whether or not the terminal user who is currently accessing is the registered terminal user based on the result of the collation judgment between the registered fingerprint and the collated fingerprint by the fingerprint collation and the result of the characteristic abnormality in the collated fingerprint. The transmission is notified to the communication terminal 40 or the net terminal which is the source of the authentication request.

As a result, whether or not the user who is currently using the communication terminal 40 is the principal is authenticated, and a transaction requiring identification can be performed.

Therefore, according to the fingerprint collation function of the fingerprint collation device having the above-described configuration, the collation fingerprint image data of the terminal user acquired by the fingerprint reading device 41 and the sensor unit 42 of the communication terminal 40 and the sensor data relating to the collation fingerprint are obtained. When is received by the collation server 30 and stored in the accumulated data memory 33, characteristics such as the pattern of the collated fingerprint and variations in reading position are determined, and a fingerprint collation algorithm suitable for this collated fingerprint characteristic is selected and determined. Then, according to the fingerprint collation processing software (program) of the fingerprint collation algorithm suitable for this collation fingerprint characteristic, the registered data memory 32 for the collation fingerprint data of the terminal user.
Since the person is authenticated by collating with the registered fingerprint data in, the more accurate and efficient authentication process can be performed when introducing the person authentication using a communication network.

(Characteristic Judgment Processing) Next, characteristic judgment processing (SA) accompanying the fingerprint registration processing at the sales terminal 10 in the fingerprint collation system of the first embodiment, the collation server in the fingerprint collation apparatus of the second embodiment. The characteristic determination processing (BA) accompanying the fingerprint authentication processing in 30 will be described.

FIG. 5 is a flow chart showing the pattern determination process which is the fingerprint characteristic determination process (No. 1).

This pattern determination processing (steps A11 to A1)
In 3), the fingerprint image read by the fingerprint reading device (15, 41) is compared with, for example, sample images of various fingerprint patterns stored in advance, so that the pattern having a high matching rate is read and the fingerprint characteristic of the fingerprint image is read. To judge.

The types of patterns are roughly classified into swirl patterns / bow patterns / hoof patterns / transform patterns. Sample fingerprints other than the variant patterns are prepared. The swirl patterns / bow patterns / hoof patterns If the collation rate of any of the sample images of the print is less than the specified value, it is determined as a transformation print.

Spiral pattern: Ring-shaped, spiral-shaped, double hoof-shaped ridges in the center of the fingerprint (a raised line for each fingerprint)
There is a fingerprint with a delta on the left and right.

Bow-shaped pattern: A fingerprint formed by bow-shaped lines. The feature is that there is no delta on the left and right of the finger.

Hoof print: A fingerprint that includes hoof lines and has a delta on the opposite side of the flow direction. Among them, the hoof-shaped crest is
The hoof-shaped line flows to the thumb side, and the B-shaped hoof pattern refers to the hoof-shaped line flowing to the little finger side.

Transformation pattern: A fingerprint that does not belong to any of a spiral pattern, a bow pattern, and a hoof pattern. The metamorphic crest is congenital, and is formed only by dots or short ridges, or is formed by ridges flowing vertically.

FIG. 6 is a flow chart showing the skin quality judgment processing which is the fingerprint characteristic judgment processing (No. 2).

This skin quality judgment processing (steps A21 to A2
In 3), the hardness and skin water content of the finger placed on the reading surface of the fingerprint reading device (15, 41) are measured by a tactile sensor or a capacitance sensor provided in the sensor unit (16, 42). , It is judged as the skin quality characteristics of fingerprints, which are roughly classified into sweat / normal / dry.

FIG. 7 is a flowchart showing the ridge line determination process which is the fingerprint characteristic determination process (No. 3).

This ridge line determination process (steps A31 to A3)
In 3), the fingerprint reading device (15, 4) is provided by the electrostatic capacity sensor provided in the sensor unit (16, 42).
The ridges of the fingerprint of the finger placed on the reading surface of 1) are discriminated, and the ridge characteristics of the fingerprint are roughly classified into high / normal / low.

FIG. 8 is a flow chart showing the age determination process which is the fingerprint characteristic determination process (No. 4).

This age determination process (steps A41 to A4
In 3), the selling terminal 10 of the first embodiment determines the age according to the personal information of the telephone purchase user input by the input / output device 14, and the collation server 30 of the second embodiment inputs the age from the communication terminal 40. The age is discriminated according to the personal information of the terminal user, and the age characteristic of the fingerprint is roughly classified into 18 years or older / 12 to 17 years old / 7 to 11 years old / 0 to 6 years old.

FIG. 9 is a flow chart showing the gender determination process which is the fingerprint characteristic determination process (No. 5).

This sex determination process (steps A51 to A5)
In 3), the selling terminal 10 of the first embodiment determines the gender according to the personal information of the telephone purchasing user input by the input / output device 14, and the collation server 30 of the second embodiment inputs the information from the communication terminal 40. The gender is determined according to the personal information of the terminal user, and the gender characteristics of the fingerprint are roughly classified into male / female.

FIG. 10 is a flow chart showing the skin color judgment processing which is the fingerprint characteristic judgment processing (No. 6).

This skin color judgment processing (steps A61 to A
63), for the fingerprint image read by the fingerprint reading device (15, 41), for example, by calculating the average gradation level of the binarized image data, the skin color characteristic of the fingerprint is roughly classified into dark / normal / light. To judge.

FIG. 11 is a flow chart showing the skin elasticity judgment processing which is the fingerprint characteristic judgment processing (No. 7).

This skin elasticity determination process (steps A71 to A71)
A73) uses the tactile sensor provided in the sensor unit (16, 42) to detect the fingerprint reading device (15, 4).
The hardness of the skin of the finger placed on the reading surface of 1) is measured and judged as the skin elasticity characteristic of the fingerprint roughly classified into hard / normal / soft.

FIG. 12 is a flow chart showing the finger size determination process which is the fingerprint characteristic determination process (No. 8).

This finger size determination processing (steps A81 to
In A83), the fingerprint sensor (15, 4) is provided by the contact sensor provided in the sensor unit (16, 42).
The size of the finger placed on the reading surface in 1) is measured and judged as the size characteristics of the finger classified into large / normal / small.

FIG. 13 is a flowchart showing the positional deviation amount judgment processing which is the fingerprint characteristic judgment processing (No. 9).

This positional deviation amount determination processing (step A91)
In (A93), the fingerprint reading device (15, 41) is caused to perform the reading operation of the fingerprint of the telephone purchasing user or the communication terminal user 10 times, the average positional deviation amount X is calculated, and it is judged as the positional deviation characteristic of the fingerprint reading. To do.

FIG. 14 is a flowchart showing the distortion amount variation determination processing which is the fingerprint characteristic determination processing (10).

The distortion amount variation is a habit of placing the finger on the reading surface of the fingerprint reading device (15, 41) for each individual. For example, a person gently placing the finger horizontally with respect to the sensor surface (reading surface). For some people, the finger tip first contacts the sensor surface and then the entire fingerprint surface. In the former case, there is almost no distortion of the fingerprint, but in the latter case, the upper side (fingertip side) of the fingerprint tends to be compressed. Even when the finger is placed horizontally, if the fingertip is slid while pressing it to increase the contact surface after touching the center of the fingerprint to the sensor surface, the center part of the fingerprint will be centered on the outer pattern. The pattern tends to move downward (the base of the finger). Such distortion of the fingerprint pattern due to how the finger is placed on the sensor surface of each individual is referred to as “distortion variation”.

This distortion amount variation determination processing (step A
101 to A103), the fingerprint reading device (15, 4
The reading operation of the fingerprint of the telephone purchasing user or the communication terminal user according to 1) is performed 10 times, the average variation Y of the reading position and the angle is calculated, and it is judged as the variation characteristic of the fingerprint reading distortion amount.

FIG. 15 is a flow chart showing the pressure variation judgment processing which is the fingerprint characteristic judgment processing (11).

This pressure variation determination process (step A1)
11 to A113), the fingerprint reading device (15, 4
The fingerprint reading operation of the telephone purchasing user or the communication terminal user according to 1) is performed 10 times, and the finger pressure is measured by the pressure sensor provided in the sensor unit (16, 42) each time, and the pressure variation Z is calculated. Therefore, it is determined as the pressure variation characteristic of fingerprint reading.

That is, according to any or a combination of the characteristics of the fingerprint itself and the reading characteristics judged by such fingerprint characteristic judgment processing (Nos. 1 to 11).
By selecting the optimal fingerprint matching algorithm and performing fingerprint matching processing according to the matching processing software (program), fingerprints with high matching performance are always provided under various finger conditions or various fingerprint reading conditions. Collation can be performed efficiently.

(Fingerprint reading / positioning mechanism) Next, the selling terminal 10 in the fingerprint collation system of the first embodiment.
The fingerprint reading positioning mechanism of each of the fingerprint reading device 15 and the fingerprint reading device 41 of the communication terminal 40 in the fingerprint collation device of the second embodiment will be described.

FIG. 16 is a diagram showing a fingerprint reading positioning mechanism (No. 1) in which the fingerprint reading surface is inclined. FIG. 16A is a view showing the external configuration of the same, and FIG. 16B is its side sectional configuration. FIG.

This fingerprint reading / positioning mechanism (1)
Then, the fingerprint reading surface 5 for mounting the fingerprint surface Fb of the finger F on the upper surface 51 of the fingerprint reading box 50.
2 is inclined downward from the front side to the back side.

Then, the front angle (tilt start angle) 5 of the fingerprint reading surface 52 in which the first joint Fa of the finger F is tilted downwardly 5
3 from the front angle (tilt start angle) 53 of the fingerprint reading surface 52 to the front end portion of the reading surface 52 so that the center of the fingerprint surface Fb is substantially at the center of the fingerprint reading surface 52. Set the distance D.

As described above, by using the fingerprint reading / positioning mechanism (No. 1) in which the fingerprint reading surface 52 is inclined, it is possible to naturally contact the reading surface 52 not from the tip of the finger F but from the belly of the finger F. In addition to obtaining a large contact area, the strain generated on the fingerprint surface Fb can be minimized without applying an excessive contact pressure.

In addition, the front angle (tilt start angle) 53 of the fingerprint reading surface 52 in which the first joint Fa of the finger F is tilted downward is described.
By positioning by touching against, it is possible to obtain stable fingerprint image data without wobbling.

FIG. 17 is a diagram showing a fingerprint reading positioning mechanism (No. 2) provided with a finger reading guide marking.

This fingerprint reading / positioning mechanism (2)
Then, with respect to the upper surface 51 of the fingerprint reading box 50, guide markings 54a, 54 for correctly setting the fingerprint surface Fb of the finger F on the fingerprint reading surface 52.
b is provided and configured.

FIG. 18 is a view showing a fingerprint reading / positioning mechanism (No. 3) provided with a finger reading guide protrusion.

This fingerprint reading / positioning mechanism (3)
Then, the upper surface 51 of the fingerprint reading box 50 is provided with guide protrusions 55a and 55b for correctly setting the fingerprint surface Fb of the finger F on the fingerprint reading surface 52.

FIG. 19 is a view showing a fingerprint reading positioning mechanism (No. 4) provided with a finger reading guide depression.

This fingerprint reading / positioning mechanism (4)
Then, with respect to the front corner of the upper surface 51 of the fingerprint reading box 50, the fingerprint surface Fb of the finger F is attached to the fingerprint reading surface 52.
Is formed by providing a guide recess 56 for correctly setting the position.

In this way, the guide markings 54a, 5a
4b, the guide protrusions 55a and 55b, and the guide depression 56 are provided in the fingerprint reading positioning mechanism (Nos. 2 to 4), so that the pressing position and the pressing angle of the finger F with respect to the fingerprint reading surface 52 can be properly regulated. it can.

FIG. 20 shows a fingerprint reading / positioning mechanism (No. 5) provided with a finger reading guide switch.

This fingerprint reading / positioning mechanism (5)
Then, with respect to the upper surface 51 of the fingerprint reading box 50, a guide switch 57 for correctly setting the fingerprint surface Fb of the finger F on the fingerprint reading surface 52 and starting the reading operation is configured.

As described above, by using the fingerprint reading / positioning mechanism (No. 5) provided with the guide switch 57, the position where the finger F is pressed against the fingerprint reading surface 52.
Not only can the pressing angle be properly regulated, but also the contact area of the fingerprint surface Fb of the finger F with the fingerprint reading surface 52 can be naturally increased in accordance with the pressing operation of the switch 57.

FIG. 21 is a view showing a fingerprint reading / positioning mechanism (No. 6) provided with a variable guide 60 corresponding to a finger size. FIG. 21A is a view showing a guide positioning state corresponding to a small-sized finger. FIG. 6B is a diagram showing a guide positioning state according to a medium-sized finger, and FIG. 7C is a diagram showing a guide positioning state according to a large-sized finger.

This fingerprint reading and positioning mechanism (6)
Then, the scissor-shaped finger guide poles 62a, 62a having the fulcrum 61 at a position apart from the center of the fingerprint reading surface 52 on one side
b is provided, and the finger guide poles 62a and 62b are provided.
A contraction spring 63 is provided at the tip on the opposite side of the fulcrum 61 from the finger guide side so that the finger guide poles 62a and 62b are always actuated in the closing direction as shown by an arrow i.

Further, between the finger guide poles 62a and 62b, between the poles 62a and 62b on the finger guide side, as shown by an arrow j, an extension spring 64 having an extension force toward the fingerprint reading surface 52 and the extension spring 64 are provided. A finger end stopper 65 attached to the tip of the spring is provided, and the finger end stopper 65 is
By the step-like stopper regulating portions 66a and 66b formed on the inner surface of each pole 62a and 62b, the pole 62
The stop position is regulated according to the opening / closing condition of a and 62b.

As described above, the variable guide 6 corresponding to the finger size is used.
In the fingerprint reading and positioning mechanism (6) provided with 0,
The finger guide width by the finger guide poles 62a and 62b becomes variable according to the size of the finger F, and accordingly, when the stop position by the finger end stopper 65 is a small size finger F, the stop position near the fingerprint reading surface 52. In addition, in the case of a large-sized finger F, the finger F is restricted to a stop position farther from the fingerprint reading surface 52.

As a result, regardless of the size of the finger F with respect to the fingerprint reading surface 52, the fingerprint surface Fb can always be guided and stabilized at appropriate positions in the front-rear direction and the left-right direction.

FIG. 22 is a view showing a fingerprint reading / positioning mechanism (No. 7) provided with a finger pressing guide. FIG. 22 (A) shows the non-guide state, and FIG. It is a figure which shows a contact guide state.

This fingerprint reading and positioning mechanism (7)
In addition, a reading box 50 'having a U-shaped cross section in which the finger F is inserted from the tip is provided, and an extension force is exerted in a direction indicated by an arrow k from an upper portion of an insertion position of the finger F corresponding to the fingerprint reading surface 52. A pressing guide 59 for the finger F is provided via a spring 58. This pressing guide 59
The contact surface with the finger F has rollers 59a and 59b that freely rotate in the insertion / removal direction of the finger F.

As described above, in the fingerprint reading / positioning mechanism (No. 7) provided with the pressing guide 59 for the finger F, the finger F
When the finger is inserted from the tip of the finger so as to correspond to the fingerprint reading surface 52, the rollers 59a and 59b of the pressing guide 59 come into rotational contact with the upper side surface (nail side) of the finger F, and the pressing guide 5
9 is pushed up against the spring force (k) of the spring 58.
As a result, the fingerprint surface Fb of the finger F can be stably pressed against the fingerprint reading surface 52 by the spring force (k), and a biased fingerprint image can be obtained.

FIG. 23 is a view showing a fingerprint reading / positioning mechanism (No. 8) having a finger pressing hook. FIG. 23 (A) shows its non-pressing state, and FIG. 23 (B) shows its finger. Is a diagram showing an initial state of insertion, and FIG. 7C is a diagram showing a pressed state of the finger.

This fingerprint reading / positioning mechanism (8)
Then, a reading box 50 'having a U-shaped cross section in which the finger F is inserted from the tip is provided, and the upper portion of the U-shaped portion is used as a finger hook 50f having a restoring force in the direction of the fingerprint reading surface 52 shown by an arrow k. Constitute.

In this way, the finger F pressing hook 50f
In the fingerprint reading / positioning mechanism (No. 8), the finger F is hooked from the tip to the finger hook 50f and inserted between the fingerprint reading surface 52 and the fingerprint of the finger F due to the restoring force (k) of the finger hook 50f. Surface Fb is fingerprint reading surface 5
It can be stably pressed against No. 2 and a fingerprint image without bias can be obtained.

FIG. 24 is a view showing a fingerprint reading / positioning mechanism (No. 9) provided with an automatic finger pressing guide.
The figure (A) is a figure which shows the non-guide state, the figure (B) is a figure which shows the initial insertion state of the finger, and the figure (C) is a figure which shows the automatic pressing guide state.

This fingerprint reading / positioning mechanism (9)
Then, a reading box 50 'having a U-shaped cross section in which the finger F is inserted from the tip is provided, and is automatically lowered by an actuator from above the insertion position of the finger F corresponding to the fingerprint reading surface 52 in a direction indicated by an arrow k. An automatic pressing guide 71 for the finger F is provided. In addition, a camera 72 for photographing the nail Fc of the inserted finger F is provided in the upper front part of the insertion position of the finger F.

The drive unit of the automatic pressing guide 71 and the camera 72 are connected to the controller of the fingerprint reading device (15, 41), and when the finger F is inserted corresponding to the fingerprint reading surface 52, the camera 72 is inserted. The insertion of the finger F is detected by the analysis of the captured image by, and the descending drive of the automatic pressing guide 71 is started. Then, when the finger F is pressed against the fingerprint reading surface 52 by the automatic pressing guide 71 and the discoloration of the nail Fc of the finger F is detected by the analysis of the image taken by the camera 72, the automatic pressing guide 71 The descent drive is stopped, and the fingerprint surface Fb of the finger F is pressed against the fingerprint reading surface 52 with an appropriate pressure.

As a result, the fingerprint surface Fb of the finger F can be stably pressed against the fingerprint reading surface 52, and a fingerprint image without bias can be obtained.

That is, a fingerprint reading device (1 provided with such a fingerprint reading positioning mechanism (Nos. 1 to 9) is provided.
5, 41), it is possible to obtain uniform and good fingerprint image data at all times, and in combination with the selection decision of the matching algorithm suitable for the fingerprint characteristics, the false rejection rate and the false acceptance rate are extremely low. , It becomes possible to build a fingerprint matching system with high matching performance.

The invention of the present application is not limited to the above-described embodiments, but can be variously modified at the stage of implementation without departing from the spirit of the invention. Furthermore, each of the embodiments includes inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituents are deleted from all constituents shown in one embodiment or constituents shown in some embodiments are combined, it is described in the column of the problem to be solved by the invention. When the problem described above can be solved and the effects described in the section of the effects of the invention can be obtained, a structure in which these constituent requirements are deleted or combined can be extracted as the invention.

[0118]

As described above, according to the fingerprint collation system according to the first aspect of the present invention, the fingerprint pattern or reading of the fingerprint of the electronic device user is accompanied by the reading of the fingerprint image of the electronic device user by the fingerprint reading means. The characteristics such as the amount of misalignment due to the above are determined, and the fingerprint matching algorithm according to the characteristics of the electronic device user is selected. Fingerprint image data of the electronic device user read by the fingerprint reading means and fingerprint matching processing software (program) according to the fingerprint matching algorithm selected according to the user characteristic are registered in the electronic device. The fingerprint collation processing software (program) registered in the device enables the fingerprint collation between the device user and the registered fingerprint.

According to the fingerprint collation apparatus of the second aspect of the present invention, when the fingerprint image of the collation target person is acquired by the fingerprint acquisition means, the fingerprint image of the collation target person is acquired along with the acquisition of the fingerprint image. The characteristics such as the print pattern or the amount of positional deviation due to reading are determined, and the fingerprint matching algorithm according to the characteristics of the matching target person is selected. Then, the collation process of the fingerprint image data acquired by the fingerprint acquisition means is performed based on the fingerprint collation processing software (program) according to the fingerprint collation algorithm selected according to the characteristic of the collation target person. .

Therefore, according to the present invention, it becomes possible to efficiently perform fingerprint matching with high matching performance by selecting matching algorithms according to various finger conditions or various fingerprint reading conditions.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a fingerprint matching system according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing a fingerprint registration process by the fingerprint matching system.

FIG. 3 is a diagram showing a configuration of a fingerprint matching device according to a second embodiment of the present invention.

FIG. 4 is a flowchart showing a fingerprint authentication process by the fingerprint matching device.

FIG. 5 is a flowchart showing a pattern determination process which is a fingerprint characteristic determination process (No. 1).

FIG. 6 is a flowchart showing skin quality determination processing that is fingerprint characteristic determination processing (No. 2).

FIG. 7 is a flowchart showing ridge line determination processing which is fingerprint characteristic determination processing (3).

FIG. 8 is a flowchart showing age determination processing which is fingerprint characteristic determination processing (No. 4).

FIG. 9 is a flowchart showing sex determination processing which is fingerprint characteristic determination processing (No. 5).

FIG. 10 is a flowchart showing skin color determination processing which is fingerprint characteristic determination processing (No. 6).

FIG. 11 is a flowchart showing skin elasticity determination processing which is fingerprint characteristic determination processing (No. 7).

FIG. 12 is a flowchart showing finger size determination processing which is fingerprint characteristic determination processing (No. 8).

FIG. 13 is a flowchart showing a positional deviation amount determination process which is a fingerprint characteristic determination process (No. 9).

FIG. 14 is a flowchart showing distortion amount variation determination processing which is fingerprint characteristic determination processing (10).

FIG. 15 is a flowchart showing pressure variation determination processing which is fingerprint characteristic determination processing (Part 11).

16A and 16B are views showing a fingerprint reading / positioning mechanism (No. 1) in which the fingerprint reading surface is tilted, in which FIG. 16A is a diagram showing an external configuration thereof, and FIG. Fig.

FIG. 17 is a diagram showing a fingerprint reading positioning mechanism (No. 2) provided with a finger reading guide marking.

FIG. 18 is a diagram showing a fingerprint reading positioning mechanism (No. 3) provided with a finger reading guide protrusion.

FIG. 19 is a diagram showing a fingerprint reading positioning mechanism (No. 4) provided with a finger reading guide depression.

FIG. 20 is a view showing a fingerprint reading positioning mechanism (No. 5) provided with a finger reading guide switch.

FIG. 21 is a diagram showing a fingerprint reading positioning mechanism (No. 6) provided with a variable guide corresponding to a finger size, and FIG. 21A is a diagram showing a guide positioning state according to a small-sized finger; FIG. 9B is a diagram showing a guide positioning state according to a medium-sized finger, and FIG. 9C is a diagram showing a guide positioning state according to a large-sized finger.

22A and 22B are views showing a fingerprint reading / positioning mechanism (No. 7) provided with a finger pressing guide; FIG. 22A is a drawing showing its non-guide state; FIG. 22B is its pressing guide. The figure which shows a state.

23A and 23B are views showing a fingerprint reading / positioning mechanism (No. 8) having a finger pressing hook, where FIG. 23A shows the non-pressing state, and FIG. 23B shows the insertion of the finger. The figure which shows an initial state, the figure (C) is a figure which shows the pressing state of the finger.

FIG. 24 is a diagram showing a fingerprint reading / positioning mechanism (No. 9) provided with an automatic finger pressing guide, and FIG.
Shows the non-guided state, FIG. 6B shows the initial insertion state of the finger, and FIG. 6C shows the automatic pressing guide state.

[Explanation of symbols]

10 ... Terminal device of mobile phone store (sales terminal) 11 ... CPU (central processing unit) 12 ... Storage device 13 ... RAM 14 ... I / O device 15 ... Fingerprint reader 16 ... Sensor unit 17 ... Writing device 20 ... Mobile phone with fingerprint matching function 30 ... Collation server 31 ... Program memory 32 ... Registered data memory 33 ... Accumulated data memory 40 ... Communication terminal 41 ... Fingerprint reader 42 ... Sensor unit F ... finger Fa ... 1st joint Fb ... Fingerprint surface Fc ... nail 50… Fingerprint reading box 50 '... U-shaped cross section fingerprint reading box 50f ... finger hook 51 ... Reading box top surface 52 ... Fingerprint reading surface 53… Front angle of reading surface (tilt start angle) 54a, 54b ... Guide marking 55a, 55b ... Guide protrusion 56… Guide depression 57… Guide switch 59… Pushing guide 60 ... Variable guide for finger size 62a, 62b ... Finger guide poles 65 ... Finger end stopper 66a, 66b ... Stopper regulation part 71 ... Automatic pushing guide 72 ... Camera

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 4C038 FF01 FF05 FG01                 5B043 AA01 AA09 BA02 CA09 DA05                       DA06 FA02 GA01 GA17                 5B047 AA25 AB02 BA02 BB04 BC16                       BC20 BC23 CA12 CB22 DC09                 5J104 AA07 KA01 KA17 PA02

Claims (9)

[Claims] 1. A fingerprint reading unit for reading a fingerprint image of an electronic device user, a characteristic judging unit for judging a characteristic of the electronic device user when the fingerprint image is read by the fingerprint reading unit, and a judgment by the characteristic judging unit. A matching algorithm selecting means for selecting a fingerprint matching algorithm according to the characteristics of the electronic device user, and a fingerprint selected by the matching algorithm selecting means together with fingerprint image data of the electronic device user read by the fingerprint reading means. A fingerprint collation system comprising: a registration unit that registers a fingerprint collation processing program according to a collation algorithm in the electronic device. 2. A fingerprint acquisition unit that acquires a fingerprint image of a verification target person, a characteristic determination unit that determines a characteristic of the verification target person when the fingerprint image is acquired by the fingerprint acquisition unit, and the characteristic determination unit. By means of the fingerprint acquisition means using a matching algorithm selecting means for selecting a fingerprint matching algorithm according to the characteristics of the judged target person, and a fingerprint matching processing program according to the fingerprint matching algorithm selected by the matching algorithm selecting means. A fingerprint collation device comprising: a fingerprint collation unit that collates the acquired fingerprint image data. 3. The fingerprint collation system according to claim 1, or the fingerprint collation apparatus according to claim 2, wherein the characteristic is a fingerprint pattern characteristic. 4. The fingerprint collation system according to claim 1, or the fingerprint collation device according to claim 2, wherein the characteristic is a characteristic of a finger skin. 5. The fingerprint collation system according to claim 1, or the fingerprint collation apparatus according to claim 2, wherein the characteristic is a characteristic of a ridge of a fingerprint. 6. The fingerprint collation system according to claim 1, or the fingerprint collation apparatus according to claim 2, wherein the characteristic is a characteristic of a finger size. 7. The fingerprint collation system according to claim 1, or the fingerprint collation apparatus according to claim 2, wherein the characteristic is a characteristic of an amount of misalignment caused by reading a fingerprint. 8. The fingerprint collation system according to claim 1, or the fingerprint collation apparatus according to claim 2, wherein the characteristic is a characteristic of distortion variation caused by reading a fingerprint. 9. The fingerprint collation system according to claim 1, or the fingerprint collation apparatus according to claim 2, wherein the characteristic is a characteristic of pressure variation caused by reading a fingerprint.