JP2004021615A - Image verification system, image verification device, information processor, image verification method, and program describing same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform image verification excellent in security and highly accurate for authentication. <P>SOLUTION: This system is provided with a cellular phone 40 comprising a sensor which acquires a partial image of an object, a control means which controls the sensor to acquire a plurality of partial images of the object a plurality of times in the prescribed order, and a communication means which can transmit a plurality of the acquired partial images; and an authentication device 43 comprising a database 45 which stores partial image information in advance to verify a plurality of the partial images acquired in the prescribed order with each of a plurality of the partial images with the acquisition order, a verification means 44 which verifies a plurality of the partial images transmitted from the cellular phone 40 on the basis of the partial image information stored in the database 45, and a communication means which receives the acquired partial images and transmits the result of the collation by the verification means to the cellular phone 40. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an image matching system, an image matching device, an information processing device, an image matching method, and a program describing the method. In particular, biometric information such as a fingerprint and authentication information such as a barcode are read and registered as partial images with a sensor. BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image collating system, an image collating device, an information processing device, an image collating method, and a program describing the method.
[0002]
[Prior art]
An image collation device for inputting an image such as a fingerprint and collating with a pre-registered image is used in a device such as a computer in which data requiring confidentiality is recorded, and an image input device for the image collation device. Area sensors and line sensors have been used.
[0003]
On the other hand, the chances of handling personal information and electronic commerce on mobile phones and PDAs are increasing. There is an increasing demand to mount the device on a portable terminal.
[0004]
[Problems to be solved by the invention]
However, when an image such as a fingerprint is collated using an area sensor, an imaging surface having an area corresponding to the size of the finger is required, so that the number of sensor chips per wafer is small, and the cost of the sensor is high. Become.
[0005]
In addition, since the size of the area sensor is large, it cannot be mounted on a portable terminal. Further, if the size is sacrificed, a required image cannot be obtained, and there is a problem that accuracy is reduced.
[0006]
On the other hand, when a line sensor is used, the size and cost are satisfactory, but it is necessary to move the subject on the line sensor so as to scan it, and acquire an image on the surface. Requires a certain level of learning from the user. There was also a problem in many cases that recognition was not possible due to instability in image acquisition.
[0007]
In addition, it has been pointed out that fingerprint authentication may erroneously recognize even a pseudo finger (fake finger) made of gelatin.
[0008]
Furthermore, fingerprint authentication has a problem that if the fingerprint information is leaked to another person, there is no other way but to change the finger.
[0009]
Furthermore, since fingerprint authentication leaves the entire fingerprint image as registration data, there is a user's psychological anxiety at the time of acquisition or when a registered device is lost.
[0010]
[Means for Solving the Problems]
An image matching system according to the present invention includes a sensor that acquires a partial image of a subject, a control unit that controls the sensor to acquire a plurality of partial images of the subject a plurality of times in a predetermined order, and a plurality of acquired plural images. An information device having communication means capable of transmitting a partial image; and
Registration means for registering in advance partial image information for comparing a plurality of partial images obtained by acquiring the partial images of the subject in a predetermined order with each of the plurality of partial images and the acquisition order; A collation unit that collates the plurality of partial images transmitted from the device based on the partial image information registered in the registration unit, and receives the acquired partial image and transmits a collation result by the collation unit to the information device. An image collating system comprising: an authentication device having a communication unit for transmitting;
[0011]
The image collating apparatus of the present invention further includes a sensor for acquiring a partial image of the subject, control means for controlling the sensor so as to acquire a plurality of partial images of the subject a plurality of times in a predetermined order, A plurality of partial images obtained in a predetermined order of the images, and a registration unit for previously registering partial image information for matching with each of the plurality of partial images and the acquisition order,
An image collating apparatus comprising: collating means for collating a plurality of partial images acquired by the sensor using the control means based on partial image information registered in the registration means.
[0012]
In the image matching method according to the present invention, a step of pre-registering partial image information for collating a plurality of partial images obtained by acquiring partial images of a subject in a predetermined order in each of the plurality of partial images and the acquisition order. When,
Acquiring a plurality of partial images of the subject a plurality of times in the predetermined order by a sensor,
Collating a plurality of partial images acquired by the sensor based on the registered partial image information.
[0013]
A program according to the present invention describes the above-described image matching method according to the present invention.
[0014]
In the present invention, the sensor is controlled so as to obtain a plurality of partial images of a subject in a predetermined order a plurality of times, and the plurality of partial images obtained in a predetermined order of the subject partial images are respectively obtained from the plurality of partial images. By comparing the partial images with the acquisition order, a plurality of partial images can be combined and encrypted.
[0015]
When the present invention is applied to a fingerprint collation device, a plurality of partial images of a fingerprint are acquired multiple times using an area sensor whose area is set to be smaller than one-fifth of a finger. Without synthesizing, by collating each with a plurality of registered partial images registered in advance, performing authentication using the plurality of collation results, and collating according to the acquisition order of the partial images of the fingerprint, Multiple partial images can be combined and encrypted.
[0016]
Further, in the present invention, security can be further enhanced by providing a means for inputting a password and registering a plurality of registered partial images in correspondence with each character (number) of the password.
[0017]
In the present invention, by using an area sensor whose area is set to be a fraction of a finger, the cost area is reduced to a fraction of the conventional size, and the area sensor can be applied to a portable terminal. . Compared to a product using a line sensor, a stable fingerprint image can be obtained, and complexity can be reduced.
[0018]
Furthermore, when the present invention is applied to a fingerprint collation device, for example, if a false finger is created and attempted to be illegally used, and if the entire finger is registered as in the prior art, it is used with a probability of 1/10. Although the fingerprint that is being identified is specified, for example, if three partial fingerprint images of an area about 1/3 of the first joint are selected from an arbitrary part of an arbitrary finger, the probability is (1). / 30) raised to the third power, and the security level can be dramatically increased. Also, if the fingerprint information has been leaked to another person, in the past it was necessary to change the finger and re-register, but in the present invention, it is only necessary to change the combination of which finger and which position, Variations are abundant and easy to change. In addition, since the entire fingerprint image is not left as registration data, psychological uneasiness of the user at the time of acquisition or loss of a registered device can be reduced.
[0019]
Furthermore, by combining with a password, the security level can be increased and a plurality of persons can be registered. Even if the password itself is plain and easy to memorize, high security can be maintained.
[0020]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
[0021]
(1st Embodiment)
In the present embodiment, an image matching system will be described in which electronic communication is performed by authenticating a fingerprint of a communication destination server with a mobile phone. In the case of extracting feature points of a fingerprint, template matching or frequency analysis may be used instead of the matching algorithm. In this embodiment, for example, a sensor of 400 pixels horizontally and 150 pixels vertically captures three pixels in a region of about one third in the longitudinal direction of the finger.
[0022]
FIG. 1 is a diagram showing a schematic configuration of an image matching system according to the present invention. FIG. 2 is a schematic diagram of an image input device provided in a mobile phone.
[0023]
In FIG. 1, reference numeral 40 denotes a mobile phone (a PDA having a communication function can also be used) as an information device (information processing device); 41, a gateway device for connecting the mobile phone 40 to a communication network 42 such as the Internet; , 43 are authentication devices connected to the communication network 42 and are composed of an authentication server 44 and an authentication database 45. The mobile phone 40 has the image input device shown in FIG. 2 (not shown in FIG. 1), and irradiates the finger with light by the LEDs 51 arranged on both sides of the finger, and the light scattered and reflected inside the finger is applied to the optical sensor 20. It enters and reads the fingerprint of the finger. Then, the mobile phone 40 reads the partial fingerprint images of the finger a plurality of times in a predetermined order, and sequentially transmits the partial fingerprint images to the authentication device 43.
[0024]
FIG. 3 is a diagram showing an example of the configuration of the authentication device 43 of FIG. The authentication device 43 receives a plurality of partial fingerprint images transmitted from the mobile phone 40 by the communication unit 31 and temporarily stores the partial fingerprint image data in the memory 32. The feature point extraction unit 33 reads out the partial fingerprint image data from the memory 32, processes the data, extracts feature points, and stores the feature points in the feature point data storage unit 34 as feature point data. As the feature point data, while keeping the scale of the coordinates the same (while keeping the fingerprint image constant), the origin of the coordinates and the direction of the coordinate axes are represented by the coordinate positions on the orthogonal coordinate axes which are temporarily arbitrarily determined each time. Then, the distance between two adjacent feature points is calculated, and this is simultaneously stored as feature point data.
[0025]
The feature point data is compared with the feature point data of the partial fingerprint image registered in advance in the registration data storage unit (which will be an authentication database) 36 by the matching unit 37. Is displayed in the form of A control unit 35 sends a control signal to each unit. If the feature point extraction unit 33 cannot extract the feature point of the partial fingerprint image, the information is sent to the mobile phone 40 via the communication unit 31 to perform the partial fingerprint image again. When the collation of the first partial fingerprint image is completed, the collation unit 37 sends information to the mobile phone 40 via the communication unit 31 to input the fingerprint image of the next part. The collation unit 37 also performs a fingerprint authentication process described later.
[0026]
Hereinafter, the fingerprint matching operation of the image matching system will be further described.
[0027]
FIG. 4 shows a display screen of the mobile phone, and FIG. 5 is a view showing a state where a finger is placed on the optical sensor. Here, the case where the partial fingerprint image is read and the password is input to further improve the accuracy of the verification will be described. However, it is not always necessary to input the password because the acquisition order of the partial fingerprint image itself functions as a kind of password. is not.
[0028]
Here, a three-character password (for example, A1B) is set. At the time of registration / authentication, after the user ID is input, the operation keys are input three times as "A", "1", and "B" as a password. Note that three partial fingerprint images of the finger are acquired in accordance with the password (A1B) input at the time of registration, and the authentication database 45 is associated with the password (A1B) as feature point data of the three partial fingerprint images. Record.
[0029]
The authentication server 44 recognizes that the input password has three digits, and prompts the mobile phone 40 to acquire the fingerprint image three times. The mobile phone 40 displays on the display screen a display such as the one shown in FIG. According to the instruction, the user of the mobile phone, as shown in FIG. 5 (a), obtains the first partial fingerprint image by using the light sensor 20 to display the middle part in the longitudinal direction of the right little finger (upper part, middle part, lower part of the finger). An image is acquired by placing the middle part of the part), and the acquired partial fingerprint image is transferred to the authentication server 44. When the partial fingerprint image is transmitted to the authentication server 44 and the feature point cannot be extracted, the message "Please again" is displayed on the mobile phone 40, and the partial fingerprint image is obtained again.
[0030]
Next, the mobile phone 40 displays “First input completed” and proceeds to the acquisition of the second fingerprint image, as shown in FIG. 4B, “Second fingerprint input; Finger position determined by sensor” Please put on the display screen. As shown in FIG. 5B, the user of the mobile phone places the middle part of the right index finger in the longitudinal direction on the optical sensor 20 to acquire an image, and transfers the acquired partial fingerprint image to the authentication server 44. Next, the mobile phone 40 displays "end of second input", and then proceeds to acquisition of a third fingerprint image, as shown in FIG. 4 (c), "input of third fingerprint: finger determined by sensor". Please put the place of "" on the display screen. As shown in FIG. 5C, as a third fingerprint image acquisition, the user of the mobile phone acquires an image by placing the lower part of the right index finger in the longitudinal direction on the optical sensor 20 and acquires the acquired partial fingerprint image. Transfer to the authentication server 44. Then, the mobile phone displays "end". 5A to 5C, 27V indicates the vertical scanning direction of the optical sensor 20, and 27H indicates the horizontal scanning direction.
[0031]
The authentication server 44 performs the following authentication operation. The feature points of the three partial fingerprint images corresponding to the password (A1B) recorded in the authentication database 45 are compared with the feature points of the three partial fingerprint images transmitted from the mobile phone, and three pieces of each are acquired. The comparison and comparison results of the images are totaled when the comparison result value of each image is at a certain level. If the value after the totaling is equal to or more than a certain value, the authentication server authenticates the user and permits electronic commerce. However, even if the totaled value is equal to or more than a certain value, if the comparison result of any one image is equal to or less than a certain value, electronic commerce is not permitted.
[0032]
The authentication operation will be described with reference to the explanatory diagram of FIG. 6. As shown in FIG. 6, feature points (black dots in the figure) of the first to third partial fingerprint images 11, 12, and 13 transmitted from the mobile phone 40 are illustrated. And the feature points (black dots in the figure) of the first to third partial fingerprint images 21, 22, and 23 recorded in the authentication database 45, respectively. Assuming that 90% is between the fingerprint image 11 and the partial fingerprint image 21, 70% is between the partial fingerprint image 12 and the partial fingerprint image 22, and 80% is between the partial fingerprint image 13 and the partial fingerprint image 23, Then it becomes 80%. If the ratio of the feature points of each partial fingerprint image is 60% or more and the ratio of the feature points of the total fingerprint image is 70% or more, it is determined that the fingerprint belongs to the registrant. Is determined to belong to the registrant. However, if the ratio of the feature points of each partial fingerprint image is less than 60% or the ratio of the feature points of the overall fingerprint image is less than 70%, it is determined that the fingerprint is not a registrant. Further, even if the ratio of the feature points of the overall fingerprint image is 70% or more, if the ratio of the feature points of one partial fingerprint image is less than 60%, it is determined that the fingerprint is not the one of the registrant.
[0033]
During the authentication operation, as shown in FIG. 4 (c), "authentication in progress" is displayed on the mobile phone 40, and when the authentication server 44 successfully authenticates the user by collating the fingerprint, the mobile phone 40 The authentication result is transmitted to 40, and "Authentication completed" is displayed. Then, electronic commerce with the mobile phone 40 can be started. If the authentication server 44 cannot authenticate the identity of the user by collating the fingerprint, the authentication result is transmitted to the mobile phone 40, and “authentication is not possible” is displayed. In this case, electronic commerce with the mobile phone 40 cannot be started.
[0034]
Next, a partial fingerprint image authentication process (authentication routine), a partial fingerprint image collation process including an authentication process (collation routine), and a partial fingerprint image registration process (registration routine) will be described with reference to flowcharts.
[0035]
FIG. 7 is a flowchart showing an authentication routine. The authentication routine is the processing already described with reference to FIG.
[0036]
When the authentication routine is started (step S101), first, n = 0 is set (initialization operation; step S102). Next, the number of digits k (the number of digits of the password, that is, the number of collation results, where k = 3) is obtained (step S103). n on the right side of n = n + 1 is set to 0, and n on the left side is set to 1 (n = 1) (step S104), and the first comparison result P1 (partial fingerprint image 11 and partial fingerprint image 21 as described above) Is read out (step S105), and it is determined whether or not the collation result P1 is 60% or more (step S106). If the comparison result P1 is less than 60%, it is determined that the authentication cannot be performed (step S110), and the authentication routine ends (step S111).
[0037]
If the comparison result P1 is 60% or more, it is determined whether n = k (here, k = 3) has been reached (step S108). If not, the process returns to step 104, and n on the right side of n = n + 1 is set to 1 Is set, and n on the left side is set to 2 (n = 2), and steps S104 to S107 are repeated. When n = k (here, k = 3) has been reached, that is, when it is determined in step S106 that all of the first to third matching results P1 to P3 are 60% or more, the total number (P1 + P2 + P3) ) Is equal to or more than a predetermined ratio 3 (K = k = 3) × 70% (step S108). In FIG. 6, it is determined whether the total number (P1 + P2 + P3) / 3 is equal to or more than a fixed ratio of 70%. Here, the total number (P1 + P2 + P3) is evaluated.
[0038]
If (P1 + P2 + P3) ≧ 210%, it is determined that authentication has been performed (step S109), and the authentication routine ends (step S111). If (P1 + P2 + P3) <210%, it is determined that authentication cannot be performed (step S110), and the authentication routine ends (step S111).
[0039]
FIG. 8 is a flowchart showing the authentication processing (authentication routine) of the partial fingerprint image.
[0040]
When the collation routine is started (step S201), first, n = 0 is set (initialization operation; step S202). Next, "Enter your ID number and password." Is displayed on the display screen of the mobile phone, and a key input is prompted. When the ID number and the password are input (step S203), the number of digits k (for example, three digits) of the password is obtained (step S204).
[0041]
Next, n of the right side of n = n + 1 is set to 0, n of the left side is set to 1 (n = 1) (step S205), and the first (n = 1) partial fingerprint is obtained by the image input device of the mobile phone 40. An image is obtained (step S206), the partial fingerprint image is transmitted to the authentication server 44, and the authentication server 44 extracts a feature point (step S207). It is determined whether or not collation is possible (step S208). If the collation is possible, the feature point of the registered partial fingerprint image corresponding to the first character of the password registered in the authentication database 45 is registered. And (Step S209). If the level cannot be verified due to the movement of the finger during image input or the like, the process returns to step S206 to prompt the user to input an image again.
[0042]
The comparison result compared in step S209 is stored in a temporary storage unit provided in the matching unit 37 (step S210). Then, it is determined whether or not the collation result of the partial fingerprint image corresponding to the password of the third digit (k = 3) is stored in the temporary storage unit (step S211), and the password corresponding to the password of the third digit (k = 3) is determined. The process returns to step S205 until the matching result of the partial fingerprint image is stored in the temporary storage unit, and the processes from step S205 to step S211 are repeated. When the verification result of the partial fingerprint image corresponding to the password of the third digit (k = 3) is stored in the temporary storage unit, the verification result of the three partial fingerprint images corresponding to the password recorded in the temporary storage unit is read. The authentication is performed by the authentication routine 213 described with reference to FIG. 7 (step S212), and the verification routine ends (step S214).
[0043]
FIG. 9 is a flowchart showing a registration process (registration routine) of a partial fingerprint image.
[0044]
When the registration routine is started (step S301), first, n = 0 is set (initialization operation; step S302). Next, "Please enter your password" is displayed on the display screen of the mobile phone, and a key input is prompted. When the password is input (step S303), the number of digits k (for example, three digits) of the password is obtained (step S304).
[0045]
Next, n of the right side of n = n + 1 is set to 0, n of the left side is set to 1 (n = 1) (step S305), and the first (n = 1) partial fingerprint is obtained by the image input device of the mobile phone 40. An image is obtained (step S306), the partial fingerprint image is transmitted to the authentication server 44, and the authentication server 44 extracts a feature point (step S307). It is determined whether or not collation is possible (step S308). If the collation is possible, data on the feature point is registered in the authentication database (storage unit) (step S309). If the level cannot be verified due to the movement of the finger during image input, the process returns to step S306 to prompt the user to input an image again.
[0046]
Then, it is determined whether or not the collation result of the partial fingerprint image corresponding to the third digit (k = 3) password is registered in the authentication database (storage unit) (step S310). The process returns to step S305 until the feature point data of the partial fingerprint image corresponding to the password is registered in the storage unit, and the processes from step S305 to step S310 are repeated. When the characteristic point data of the partial fingerprint image corresponding to the third digit (k = 3) password is stored in the authentication database (storage unit), the password is recorded in the authentication database (storage unit) (step S311). Then, the registration routine ends (step S312).
[0047]
The user (user) of the mobile phone only needs to store the finger position corresponding to the three digits of the password. It is not always necessary to input a password. This is because acquiring the first to third fingerprint images has the same effect as using a three-digit password. However, by inputting the password, if the password is incorrect, it is not necessary to perform the collation, so that unnecessary collation processing can be eliminated. The security level is a three-digit password + fingerprint, which is a high level.
[0048]
Since the authentication server only needs to register the feature point data of the three partial fingerprint images corresponding to the password, the management of the database is facilitated. Further, since the password and the fingerprint are combined, the security strength can be increased.
[0049]
On the other hand, the cost of the sensor is one third or less of 400 pixels in the horizontal direction and 450 pixels in the vertical direction. In particular, as the chip size of the semiconductor is reduced, the yield is improved, so that the cost is further reduced.
[0050]
In addition, since it is only necessary to change the area three times to obtain an image, a stable image can be obtained without difficulty, and the user's load is much less than when a finger slides on the sensor at a constant speed like a line sensor. And the stability of image acquisition can be improved.
[0051]
In the above-described embodiment, the image collating system in the case where the mobile phone authenticates the fingerprint to the server of the communication destination and performs the electronic commerce is described. However, the image collating device having the same function as the image collating system is used. The same function can be realized by an information processing device such as a mobile phone or a PDA having the same. In this case, the image matching device shown in FIG. 10 may be mounted on a mobile phone or a PDA. The configuration shown in FIG. 10 replaces the communication unit 31 of FIG. 3 with the image input unit 30 to form an image matching device, and this image matching device is mounted on an information processing device such as a mobile phone or a PDA.
[0052]
(Second embodiment)
A case where the PDA is authenticated for the right to access the personal information inside the PDA will be described. The basic operation is the same as in the first embodiment. For example, four partial fingerprint images are acquired by a sensor having 200 pixels horizontally and 200 pixels vertically.
[0053]
At this time, a four-character password (for example, C23D) is also set.
[0054]
At the time of registration / authentication, after pressing the operation key C, an image of the middle part in the longitudinal direction of the right thumb is obtained. Then, after pressing the operation key 2, the image of the middle part in the longitudinal direction of the right thumb is obtained again. After pressing the operation key 3, the image of the upper part in the longitudinal direction of the right thumb is acquired again, and after pressing the operation key D, each image of the middle part in the longitudinal direction of the left index finger is acquired.
[0055]
The user need only memorize the eight actions of operating the operation keys and placing each finger on the sensor, and the password may be four digits.
[0056]
However, the security level is a four-digit password + fingerprint, which is a high level.
[0057]
In addition, the divided fingerprint image is acquired in the interval between the pressing operations of the operation keys, so that the user is strongly conscious of the operation of pressing the area of each divided finger against the sensor unit. The probability that the position of the finger is placed so as to be increased.
[0058]
Furthermore, since it is sufficient to register the divided image corresponding to the password, the management of the database becomes easy, and the confidentiality of the data is increased because the fingerprint information stored in the main body is not a single screen.
[0059]
On the other hand, the cost for a sensor is 1/4 or less of 400 pixels horizontally and 400 pixels vertically. In particular, as the chip size of the semiconductor is reduced, the yield is improved, so that the cost is further reduced.
[0060]
In addition, since it is only necessary to change the position of the finger four times to acquire an image, a stable image can be acquired without difficulty, and the user's load is far more than a finger sliding on the sensor at a constant speed like a line sensor And the stability of image acquisition can be improved.
[0061]
Feature points are extracted for each region without combining the four images acquired in the memory inside the terminal. On the other hand, in a database in the PDA, feature point data of each area is registered for each character of the password. Inside the PDA, the feature points are collated for each area, and the entire fingerprint authentication is performed based on the collation results of the four areas.
[0062]
After the feature points are extracted from the four acquired images, they are compared with feature point data that is registered partial fingerprint data. The comparison and collation results of each of the four acquired images are totaled when the collation result value of each image is at a certain level, and if the value after the totaling is a certain value or more, the microcomputer inside the mobile terminal authenticates itself. Allow changes to personal information.
[0063]
Japanese Patent Application Laid-Open Nos. Hei 10-143663 (Hamamatsu Photonics) and Japanese Patent Laid-Open Publication No. Hei 7-306941 (NEC Corporation) disclose a method of reading and synthesizing a plurality of partial fingerprint images of the same finger to create overall fingerprint data. However, the present invention does not combine the acquired partial fingerprint images and collate them with the registered fingerprint data, but collates the acquired partial fingerprint images with the previously registered partial fingerprint data without combining them. I do. For this reason, the acquired partial image does not need to be converted into the entire image data, so that the user does not need to be aware that the common part is connected to each image in order to combine the partial images. Images can be obtained easily and quickly.
[0064]
In addition, since the plurality of partial images need not be each part of the same finger, the user can arbitrarily change the combination of the finger and its imaging part, and as described above, fingerprint information is leaked to another person. In the case of a mistake, for example, it is easy to change by changing the combination of the finger and the position, and the psychological anxiety of the user can be reduced. Furthermore, with such a configuration, the confidentiality of the fingerprint data itself is also increased because there are many combinations of the finger and the position as compared with the case where fingerprint data of one finger is registered. Looking at the data on the registration side, the entire fingerprint data is not registered, and damage can be reduced even if the data on the server side or information leaks from the registered device. Furthermore, by combining passwords, etc., the password can maintain a sufficiently effective security effect even if the number of digits is reduced, making it easy to memorize.On the other hand, fingerprint data can be managed in association with the password, so the registration side manages it Convenience that can be easily performed can be obtained.
[0065]
The authentication device 43 described in FIG. 3 can be configured by a computer as shown in FIG. The partial image information received by the modem 127 serving as the communication unit in FIG. 1 is stored in a memory 122 such as a DRAM, and the CPU 121 executes a program describing the flowchart (the flowchart on the authentication device side) described with reference to FIGS. (Here, the program is stored in the disk device 126.) The feature points of the partial fingerprint image are extracted, and the feature point data is stored in a storage means of the disk device 126 such as an MO and a hard disk. Then, it is compared with the previously registered partial fingerprint image information. Reference numeral 124 denotes an input unit for inputting data such as a keyboard, and 125 denotes a display unit such as a CRT for displaying a result of the comparison. Reference numeral 123 denotes a bus such as a data bus. The information processing apparatus described in FIG. 10 may be configured by a computer as shown in FIG.
[0066]
In addition, as a fingerprint detection method, a finger is irradiated with light, and the reflected light or transmitted light (or scattered / transmitted light) is detected by an optical sensor to detect the fingerprint. There is a semiconductor sensor system in which a fingerprint is detected by detecting pressure, capacity, and temperature on a semiconductor sensor such as a capacitance detection element or a temperature detection element. However, any detection method may be used. The subject for which image matching is performed includes, for example, authentication information such as a barcode in addition to biometric information such as a fingerprint.
[0067]
【The invention's effect】
As described above, according to the present invention, it is possible to perform image matching with excellent security and high authentication accuracy.
[Brief description of the drawings]
FIG. 1 is a diagram showing a schematic configuration of an image matching system according to the present invention.
FIG. 2 is a schematic diagram of an image input device provided in a mobile phone.
FIG. 3 is a block diagram showing an example of a configuration of the authentication device 43 of FIG.
FIG. 4 is a diagram showing a display screen of a mobile phone.
FIG. 5 is a diagram showing a state where a finger is placed on the optical sensor.
FIG. 6 is an explanatory diagram showing an authentication operation.
FIG. 7 is a flowchart illustrating an authentication routine.
FIG. 8 is a flowchart showing an authentication process (authentication routine) of a partial fingerprint image.
FIG. 9 is a flowchart illustrating a registration process (registration routine) of a partial fingerprint image.
FIG. 10 is a block diagram illustrating a configuration example of an image matching device.
FIG. 11 is a schematic configuration diagram of a computer constituting the authentication device described in FIG. 3 or FIG.
[Explanation of symbols]
31 Communication unit
32 memories
33 Feature point extraction unit
34 Feature point data storage
35 control unit
36 Registration data storage
37 Collation unit
38 Display
40 mobile phone
41 Gateway device
43 Authentication device
44 Authentication server
45 Authentication Database

Claims (8)

A sensor for acquiring a partial image of a subject, a control unit for controlling the sensor so as to acquire a plurality of partial images of the subject in a predetermined order, and a communication unit capable of transmitting the acquired plurality of partial images And an information device comprising:
Registration means for registering in advance partial image information for comparing a plurality of partial images obtained by acquiring the partial images of the subject in a predetermined order with each of the plurality of partial images and the acquisition order; A collation unit that collates the plurality of partial images transmitted from the device based on the partial image information registered in the registration unit, and receives the acquired partial image and transmits a collation result by the collation unit to the information device. An image collating system comprising: an authentication device having a communication unit for transmitting; 2. The image matching system according to claim 1, wherein the information device has a unit for inputting a password, the predetermined order is set in correspondence with the password, and the partial image information registered in the registration unit is An image matching system characterized by being registered in correspondence with a password. A sensor for acquiring a partial image of the subject, control means for controlling the sensor so as to acquire a plurality of partial images of the subject a plurality of times in a predetermined order, and a plurality of control means for acquiring the partial images of the subject in a predetermined order. Having partial image information for previously registering partial image information for collating with each of the plurality of partial images and the acquisition order thereof,
An image matching apparatus comprising: a matching unit that matches a plurality of partial images acquired by the sensor using the control unit based on the partial image information registered in the registration unit. 4. The image collating apparatus according to claim 3, further comprising means for inputting a password, wherein the predetermined order is set in correspondence with the password, and the partial image information registered in the registration means is associated with the password. An image collating apparatus characterized by being registered. An information processing device comprising the image matching device according to claim 3. 6. An information device according to claim 5, wherein the information processing device is a mobile phone or a PDA. A step of pre-registering partial image information for comparing a plurality of partial images obtained in a predetermined order of the partial images of the subject with each of the plurality of partial images and the acquisition order;
Acquiring a plurality of partial images of the subject a plurality of times in the predetermined order by a sensor,
Collating a plurality of partial images acquired by the sensor based on the registered partial image information. A program describing the image matching method according to claim 7.

Images