JP2005258555A - Image collating unit, image collation processing program and image collation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image collating device for preventing image collation from becoming difficult, without generating wrong decisions such as denial by the principal, even when a fingerprint image changes, due to the fact that the finger being the target of registration/collation is the dominant finger. <P>SOLUTION: The fingerprint of an index finger being the dominant finger is registered as first registration fingerprint image data Gr1, and the fingerprint of the ring finger is registered as second registration fingerprint image data Gr2; and when it is decided that the registration fingerprint image data Gr1 is matched with the collation fingerprint image data Gs1 of the first finger(index finger) of the person to be collated, the first collation fingerprint image Gr1 is registration-updated as the latest first registration fingerprint image Gr1 (steps B1 to B4). Then, when it is decided that the registration fingerprint image Gr1 of first finger(index finger) does not match the collation fingerprint image Gs1, the collation processing of the registration fingerprint image Gr2 of the second finger(medicinal finger) and the collation fingerprint image Gs2 is executed; and when it is decided that they are matched by the collation of the second finger(index finger), the first collation fingerprint image Gr1 is registration-updated as the secularly changed latest first registration fingerprint image Gr1 I(steps B5 to B10). <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image collation apparatus, an image collation processing program, and an image collation method for collating fingerprint image data, for example.

  In recent years, for example, in an entrance / exit management system for an important facility or an access permission system for various information devices, a system for performing personal authentication based on a biometric image such as a fingerprint image, an iris image, or a face image has been widely introduced. It has become.

  In such a personal authentication system, for example, in the case of a fingerprint image collation device, first, a fingerprint image of a specific finger (for example, an index finger) of a regular person read by an image sensor or the like is stored in advance as a registered fingerprint image. . Then, by using the fingerprint image of the specific finger of the target person read in the same manner for the personal authentication as a collation fingerprint image, the collation similarity with the registered fingerprint image stored in advance is calculated, and thereby the registered fingerprint image and the collation fingerprint The identity with the image is determined, and whether or not the user is the registrant is authenticated.

  On the other hand, for fingerprints, the fingerprint pattern itself is basically the same pattern even with age, but due to changes in body shape such as aging and deterioration of the skin, fatness and thinning, etc. The fingerprint image read by the image sensor undergoes changes such as the crest of the fingerprint pattern being crushed with time. In particular, when the finger is a dominant finger such as an index finger, the fingerprint surface is rubbed and the change may be remarkable.

  For this reason, if the registered fingerprint image is a fingerprint image that has been initially registered, even if it is a collation fingerprint image obtained by reading the registrant's own fingerprint after a change over time, it is not determined as the same fingerprint and is difficult to collate Become.

Therefore, when the registered fingerprint image matches the registered fingerprint image and the authentication of the registrant is obtained, the collated fingerprint image at that time is stored and updated as a new registered fingerprint image each time the matching is performed. Thus, a fingerprint image collation apparatus that prevents an erroneous determination such as rejection of the person due to a change with time has been considered (see, for example, Patent Document 1).
Japanese Patent Laid-Open No. 11-25268

  However, each time a collation fingerprint image collates with the registered fingerprint image, the conventional fingerprint image collation apparatus in which the collation fingerprint image at that time is stored and updated as a new registered fingerprint image is registered relatively frequently. In a state where fingerprint verification is performed by the user himself / herself, there is little change with time in the verification fingerprint, and although smooth fingerprint verification processing and registration fingerprint image update processing are repeated, fingerprint verification by the registrant is frequently performed. If there is not, the change with time of the verification fingerprint becomes large in the meantime, and there is a problem that collation is difficult due to erroneous determination such as rejection of the person before the registered fingerprint image is updated.

  In addition, when the registered / verified finger is changed temporarily due to a change in season, temperature, rough skin, scratches, etc., if the update process of the registered fingerprint image is performed accordingly, When the state returns to the original state, there is a problem in that collation is difficult due to erroneous determination such as rejection of the person.

  The present invention has been made in view of the above-described problems. For example, when a fingerprint image is changed due to the finger to be registered / verified being a dominant finger, or temporarily changed to a fingerprint image due to an external factor. Provided are an image collation apparatus, an image collation processing program, and an image collation method capable of preventing a collation difficult without causing an erroneous determination such as rejection of a person even when a change occurs or returns. For the purpose.

  The first image collation apparatus according to the present invention comprises a plurality of registered image storage means for storing images of a plurality of parts of a person to be registered and an image reading means for reading a partial image of the person. To determine whether the registered image of the specific part of the registered images of the plurality of parts stored by the multiple registered image storage means matches or does not match the collation image of the specific part read by the image reading means, When it is determined that there is no match, the second collation judgment unit compares the registered image of the other part stored by the plurality of registered image storage units and the collation image of the other part read by the image reading unit. Judgment of match or non-match of. When the first collation determining unit determines that the registered image of the specific part stored by the multiple registered image storage unit matches the collated image of the specific part read by the image reading unit, or the second If the matching determination unit determines that the registered image of the other part stored by the plurality of registered image storage units matches the verification image of the other part read by the image reading unit, the specific part The registered image update storage means updates and stores the specific portion collation image read by the image reading means as the specific portion registration image stored by the plurality of registered image storage means.

  According to this, even if the state of the specific part of the person to be registered is changed over time due to high usage frequency, the registered image of the specific part matches the read / collated image of the specific part at that time. Each time a determination is obtained, not only can the read collation image be updated as the latest registered image, but even if the registration image of the specific portion and the read collation image are determined to be inconsistent, the usage frequency is low. Therefore, it is possible to prevent the rejection of the person by matching the registered image of the other part with little change with time and the reading collation image, and to read the specific part determined to be non-matching according to the collation matching judgment in the other part. The verification image can be updated as the latest registered image.

  The second image collating apparatus according to the present invention includes first registered image storage means for storing an image of a specific part of a person to be registered and image reading means for reading the partial image of the person. Means for determining a match or non-match based on the similarity between the registered image of the specific part stored by the first registered image storage means and the collation image of the specific part read by the image reading means, and the match When the similarity of determination is equal to or greater than a predetermined value, the first registered image update storage unit updates the specific portion of the collation image as the registered image of the specific portion stored in the first registered image storage unit. Remember. Even if it is determined by the first collation determining means that the registered image of the specific part stored by the first registered image storage means matches the collated image of the specific part read by the image reading means, the match If the determination similarity is less than the predetermined value, the second registered image storage means stores the specific portion of the collation image as the second registered image. When the non-matching between the registered image of the specific part stored by the first registered image storage unit and the verification image of the specific part read by the image reading unit is determined by the first verification determining unit Determines whether the second registered image of the specific part stored by the second registered image storage means matches or does not match the collated image of the specific part by the second collation determining means.

  According to this, every time a match determination with a high degree of similarity between the first registered image of the specific part and the read collation image is obtained, the read collation image can be updated as the latest first registered image. Even if a match determination of the read collation image is obtained due to a change in the time, etc., if the similarity is not high, the collation image at that time is stored as the second registered image of the specific part, In a state in which the specific portion is temporarily changed, the rejection of the person can be prevented by determining the match between the second registered image of the specific portion and the read collation image. Moreover, when the temporal change of the specific portion is restored, the rejection of the person can be prevented by determining whether the first registered image matches the read collation image again.

  According to the first image collation apparatus (image collation processing program) (image collation method) according to the present invention, even if a change with time occurs due to the high usage frequency of the state of the specific part of the person to be registered, Each time a match determination between the registered image of the specific part and the reading / collating image of the specific part is obtained, the reading / collating image can be updated as the latest registered image. Even if the read collation image is determined to be non-matching, the rejection of the person can be prevented by the match determination between the registered image of the other part that is less likely to change over time due to the low usage frequency and the read collation image. In response to the collation match determination at the part, the read collation image of the specific part determined to be non-matching can be updated as the latest registered image.

  According to the second image matching apparatus (image matching processing program) (image matching method) according to the present invention, every time a match determination with a high degree of similarity between the first registered image of the specific portion and the read matching image is obtained. In addition to being able to update the read collation image as the latest first registered image, if a match determination of the read collation image is obtained due to a temporal change of a specific part, etc., if the similarity is not high, By storing the collation image as the second registered image of the specific portion, the second registered image of the specific portion matches the read collation image in a state where the temporal change of the specific portion continues. Judgment can prevent rejection. In addition, when the temporal change of the specific portion is restored, it is possible to prevent the rejection of the person by matching again between the first registered image and the read collation image.

  Therefore, according to the present invention, for example, when the fingerprint image is changed due to the finger being registered / verified, or when the fingerprint image is temporarily changed due to an external factor, or returned. However, it is possible to provide an image collation apparatus, an image collation processing program, and an image collation method that can prevent collation difficulties without causing erroneous determination such as rejection of the identity.

  Embodiments of the present invention will be described below with reference to the drawings.

(First embodiment)
FIG. 1 is a block diagram showing a configuration of an electronic circuit of a fingerprint image matching apparatus according to an embodiment of an image matching apparatus of the present invention.

  This fingerprint image collation device is composed of a control unit (CPU) 11, a storage device 12, a RAM 13 (23), an image reading device 14, a display unit 15, an input unit 16, a storage medium reading unit 18, and a transmission control unit 19. They are connected to each other via a bus 17.

  In response to an input signal from the input unit 12, the control unit (CPU) 11 is a control program stored in advance in the storage device 12, or a storage medium such as a floppy disk drive FDD from an external storage medium 18a such as a floppy disk FD. The fingerprint collation processing program read into the storage device 12 via the reading unit 18 or the storage device 22 of the external computer terminal (program server 21) from the storage device 12 via the communication network 20 to the storage device 12 via the transmission control unit 19. The read fingerprint collation processing program is activated, and the operation of each part of the circuit is controlled by using the RAM 13 (23) as a work memory.

  The storage device 12 is constituted by a ROM of a semiconductor memory, a hard disk device HDD, or the like. In addition to storing the above-described control program read by the control unit (CPU) 11 immediately after turning on the power of the verification device, the storage device 12 reads an image. The registered fingerprint image read by the device 14 (for example, the first registered fingerprint image Gr1 of the index finger having a high use frequency and the second registered fingerprint image Gr2 of the ring finger having a low use frequency) is stored.

  The RAM 13 (23) is a work memory used by the control unit (CPU) 11 when executing the above-described control program.

  The image reading device 14 is a device for acquiring registered fingerprint images Gr1 and Gr2 and collation fingerprint images corresponding thereto (for example, the first collation fingerprint image Gr1 of the index finger and the second collation fingerprint image Gs2 of the ring finger). An image sensor such as an image scanner device or a CCD (Charge Coupled Device).

  The display unit 15 is a display device such as a CRT or a liquid crystal display that displays a result of collation between the registered fingerprint image (Gr1 / Gr2) and the collation fingerprint image (Gs1 / Gs2).

  The input unit 16 is an input device such as a keyboard device for the user of the verification device to instruct the control unit (CPU) 11 to issue a fingerprint image acquisition instruction or start a fingerprint verification process described later.

  Hereinafter, in the first embodiment, the first registered fingerprint image Gr1 is described as an index finger having a high use frequency, and the second registered fingerprint image Gr2 is described as a ring finger having a low use frequency.

  FIG. 2 is a diagram showing a data memory secured in the RAM 13 of the fingerprint image matching device according to the first embodiment of the present invention.

  The RAM 13 includes a read image memory 13a, a first registered image memory 13b, a first collation image memory 13c, a second registration image memory 13d, a second collation image memory 13e, a collation similarity memory 13f, a collation result memory 13g, and a non-match. The number memory 13h and the like are secured.

  In the read image memory 13a, fingerprint image data read by the image reading device 14 is temporarily stored as a multi-tone image.

  In the first registered image memory 13b, the first (forefinger) fingerprint image data Gr1 of the registrant read by the image reading device 14 to be stored and registered in the storage device 12 is subjected to fingerprint registration processing (see FIG. 3). And stored as a first fingerprint image to be registered.

  The first collation image memory 13c stores first (index finger) fingerprint image data Gs1 to be collated read by the image reading device 14.

  In the second registered image memory 13d, the second (ring finger) fingerprint image data Gr2 of the registrant read by the image reading device 14 to be stored and registered in the storage device 12 is subjected to fingerprint registration processing (see FIG. 3). And stored as a second fingerprint image to be registered.

  In the second collation image memory 13e, second (index finger) fingerprint image data Gs2 to be collated read by the image reading device 14 is stored.

  In the matching similarity memory 13f, matching is performed on the registered fingerprint image calculated in accordance with each fingerprint matching process (see FIG. 4) of the registered fingerprint image data (Gr1 / Gr2) and the matching fingerprint image data (Gs1 / Gs2). The similarity of fingerprint images is stored.

  In the verification result memory 13g, the registration that is evaluated and determined according to the similarity between the registered fingerprint image data (Gr1 / Gr2) and the verification fingerprint image data (Gs1 / Gs2) calculated and stored in the verification similarity memory 13f. The determination result of match / non-match between the fingerprint image (Gr1 / Gr2) and the verification fingerprint image (Gs1 / Gs2) is stored.

  In the non-matching frequency memory 13h, a non-matching determination result is obtained with each fingerprint matching process (see FIG. 4) between the registered fingerprint image data (Gr1 / Gr2) and the matching fingerprint image data (Gs1 / Gs2). The number of times n is counted and stored.

  In the fingerprint image collation apparatus of the first embodiment, when the first registered index image data Gr1 and the collation fingerprint image data Gs1 are collated and a match is determined, the first collation fingerprint image Gs1 is obtained. Is updated as the first registered fingerprint image Gr1. When the first finger (index finger) is determined to be inconsistent due to the influence of the index finger, for example, the second (ringing finger) registered fingerprint image data Gr2 and the verification fingerprint image data Gs2 If the match is determined by performing the above verification, the first verification fingerprint image Gs1 is determined to be the first registered fingerprint image Gr1 that has changed over time and stored and updated.

  Next, the fingerprint registration / collation function of the first embodiment in the fingerprint image collation apparatus having the above configuration will be described in more detail.

(Fingerprint registration process)
FIG. 3 is a flowchart showing a fingerprint registration process for registering the fingerprint image of the authorized user by the fingerprint image matching device according to the first embodiment of the present invention.

  When the image reader 14 reads the first (index finger) fingerprint image of the legitimate user and stores it in the read image memory 13a in the RAM 13 (step A1), the first (index finger) read image is the first image. The first registered fingerprint image data Gr1 is stored in the registered image memory 13b (step A2).

  Subsequently, similarly, when the image reader 14 reads the second (ringing finger) image of the legitimate user and stores it in the read image memory 13a in the RAM 13 (step A3), the second (ringing finger) The read image is stored as second registered fingerprint image data Gr2 in the second registered image memory 13d (step A4).

(Fingerprint verification process)
FIG. 4 is a flowchart showing fingerprint collation processing by the fingerprint image collation apparatus according to the first embodiment of the present invention.

  When the first (index finger) fingerprint image of the person to be verified is read by the image reading device 14 and stored in the read image memory 13a in the RAM 13, the read image is stored in the first verification image memory 13c with the first verification fingerprint. It is stored as image data Gs1 (step B1).

  Then, the similarity between the first (forefinger) registered fingerprint image data Gr1 stored in the first registered image memory 13b and the first verification fingerprint image data Gs1 stored in the first verification image memory 13c is calculated. Thus, a collation determination process is performed (step B2).

  Here, when the similarity between the first (forefinger) registered fingerprint image Gr1 and the first collation fingerprint image Gs1 is equal to or higher than a preset threshold value, and the collation match is determined as the first (forefinger) fingerprint of the registrant. (Step B3), the first verification fingerprint image data Gs1 is stored and updated as the latest first registered fingerprint image data Gr1 (Step B4).

  On the other hand, for example, the first registered fingerprint image Gr1 is read this time due to the long period of time from the previous fingerprint collation to the current fingerprint collation and the large change over time of the first finger (index finger) of the registrant. If the similarity with the first collation fingerprint image Gs1 is less than a preset threshold value and a non-coincidence determination is made (step B3 (No)), the first collation fingerprint image for the first registration fingerprint image data Gr1 is determined. The number n of non-match determinations associated with the comparison with the data Gs1 is counted and stored in the non-match number memory 13h in the RAM 13 (step B5), and the number n of non-match determinations by the first finger (index finger) is previously determined. It is determined whether or not the set number of non-matches n1 has been reached (step B6).

  Here, when it is determined that the non-matching determination number n by the first finger (index finger) has reached a preset non-matching number n1 (for example, three times), the image reading device 14 then applies The second fingerprint image of the collation person is read and stored in the read image memory 13a in the RAM 13, and this read image is stored in the second collation image memory 13e as the second collation fingerprint image data Gs2 ( Step B6 → B7).

  Then, the similarity between the second (ringing finger) registered fingerprint image data Gr2 stored in the second registered image memory 13d and the second verification fingerprint image data Gs2 stored in the second verification image memory 13e is calculated. Then, the collation determination process is performed (step B8).

  Here, the second finger (ringing finger), for example, has a low frequency of use and hardly changes over time, so that the degree of similarity between the second (ringing finger) registered fingerprint image Gr2 and the second collation fingerprint image Gs2 is determined in advance. When the matching threshold is determined as the second (ringing finger) fingerprint of the registrant who is equal to or greater than the set threshold (step B9), the first matching that has been determined to be non-matching stored in the first matching image memory 13c. The fingerprint image data Gs1 is stored and updated as the latest first registered fingerprint image data Gr1 that has changed over time (step B10).

  As a result, for example, even if the fingerprint image is subject to frequent use and the fingerprint image is subject to significant changes over time, it will not be difficult to collate due to the refusal of the person. The fingerprint image of the dominant finger can be registered and updated as the correct first registered fingerprint image Gr1 of the registrant.

  On the other hand, following the non-matching determination between the first (index finger) registered fingerprint image Gr1 and the verification fingerprint image Gs1, the similarity between the second (ringing finger) registered fingerprint image Gr2 and the verification fingerprint image Gs2 is also set in advance. If it is less than the threshold value and a non-coincidence determination is made (step B9 (No)), the number n of non-coincidence determinations associated with the collation of the second registered fingerprint image data Gr2 with the second collation fingerprint image data Gs2 is also made. The count is stored, and it is determined whether or not a preset number of non-matches n2 has been reached (step B11).

  If it is determined that the non-matching determination number n by the second finger (ringing finger) has also reached a preset non-matching number n2 (for example, three times), the person to be checked this time is the registrant himself / herself. It is determined that there is not, and the series of collation processes is ended.

  Therefore, according to the fingerprint registration / collation function of the first embodiment in the fingerprint image collation apparatus having the above-described configuration, the fingerprint of the index finger, which is a dominant finger, is registered as the first registration fingerprint image data Gr1, and the fingerprint of the ring finger is secondly registered. When it is registered as registered fingerprint image data Gr2 and matched with the verification fingerprint image data Gs1 of the first finger (index finger) of the person to be verified, the first verification fingerprint image Gr1 is the latest first registered fingerprint image Gr1. As a registration update. When it is determined that the registered fingerprint image Gr1 of the first finger (index finger) does not match the verification fingerprint image Gs1, the verification process of the registered fingerprint image Gr2 and verification fingerprint image Gs2 with the second finger (ringing finger) When the match is determined by collation with the second finger (ring finger), the first collation fingerprint image Gr1 is registered and updated as the latest first registered fingerprint image Gr1 that has changed over time.

  As a result, for example, even when the fingerprint image has a clear change over time due to the frequently used finger being registered / verified, it can be prevented that collation is difficult due to the rejection of the person, The registration of the dominant finger after the change with time can be updated as the correct first registered fingerprint image Gr1 of the registrant.

  In the fingerprint collation process (see FIG. 4) of the first embodiment, the registration fingerprint image Grn and the collation fingerprint image Gsn are judged to have reached the preset number nm of the number of non-matches n, and the registration is performed. : Matching fingerprint images (Grn: Gsn) are determined to be completely non-matching, but each match / non-matching determination process includes not only the similarity between the registered fingerprint image Grn and the matching fingerprint image Gsn, but also the fingerprint. Considering the vertical / horizontal shift and rotational shift of the entire fingerprint image affected by the image reading error, it is determined as non-matching when the similarity when the vertical / horizontal shift or rotational shift is below a certain amount is less than a threshold value. It is good also as a structure.

  Further, in the first embodiment, even if the first verification fingerprint image data Gs1 is determined to match the first (index finger) registered fingerprint image data Gr1, or the non-match determination is made, the second (ring finger) And the second fingerprint image data Gs2 is determined to match the registered fingerprint image data Gr2, and the first fingerprint image data Gs1 is updated and registered as the latest first fingerprint image data Gr1. However, the second registered fingerprint image data Gr2 is also updated as the latest second registered fingerprint image data Gr2 by obtaining a match determination with the second verified fingerprint image data Gs2 for the second registered fingerprint image data Gr2. It is good also as a structure to register.

  In the first embodiment, two fingerprint images of the first (index finger) registered fingerprint image data Gr1 and the second (ringing finger) registered fingerprint image data Gr2 are registered images, and the first registered fingerprint image data Gr1 and its registered fingerprint image data Gr1 When the collation fingerprint image data Gs1 does not match, the second registration fingerprint image data Gr2 and the collation fingerprint image data Gs2 are collated. When the collation fingerprint image data Gs1 is matched, the first registration fingerprint image data Gr1 is However, the image is not limited to these two fingerprint images, but more than three fingerprint portions such as the middle finger and the little finger are also registered in the third registered fingerprint image data Gr3, 4 is registered as registered fingerprint image data Gr4,..., And when the first registered fingerprint image data Gr1 which is a specific part and the collation fingerprint image data Gs1 are not matched, any one of the second and subsequent registration fingers is registered. The image data Grn and its collation fingerprint image data Gsn are collated, and if they match, the first registered fingerprint image data Gr1 as the specific portion is replaced with the collation fingerprint image data Gs1 at that time and updated and stored. It is good also as composition to do.

  In the fingerprint registration / collation function of the first embodiment, the registered fingerprint image Gr1 of the first finger (index finger) and the registered fingerprint image Gr2 of the second finger (ringing finger) that is a reserve are stored and registered in advance. If the collation fingerprint image Gs1 with the finger (index finger) becomes non-matching due to the influence of changes over time, the collation processing with the collation fingerprint image Gs2 with the second second finger (ringing finger) is performed, and the person is rejected Although the configuration is such that erroneous determination is prevented, the storage update of the first registered fingerprint image Gr1 is very similar to the verified fingerprint image Gs as described in the fingerprint registration / verification function of the second embodiment below. Even if the matching fingerprint image Gs is determined to match, for example, if the similarity is not so high due to a slight fingerprint change, the first registered fingerprint image Gr1 is not updated and the matching fingerprint image at that time is not updated. Gs is the second registered fingerprint To register as an image Gr2. Then, in a state where the first registered fingerprint image Gr1 is not matched due to a temporary fingerprint change, the fingerprint change is restored by obtaining a matching match determination based on the second registered fingerprint image Gr2. Even in such a case, a configuration may be adopted in which a match determination can be obtained again based on the first registered fingerprint image Gr1.

(Second Embodiment)
FIG. 5 is a diagram showing a data memory secured in the RAM 23 of the fingerprint image collating apparatus according to the second embodiment of the present invention.

  In the RAM 23, a read image memory 23a, a first registered image memory 23b, a second registered image memory 23c, a collation image memory 23d, a collation similarity memory 23e, a collation result memory 23f, and the like are secured.

  In the read image memory 23a, fingerprint image data read by the image reading device 14 is temporarily stored as a multi-tone image.

  In the first registered image memory 23b, the fingerprint of a specific finger (for example, index finger) of the registrant read by the image reading device 14 to be stored and registered in the storage device 12 in accordance with the fingerprint registration processing (see FIG. 6). Image data is stored and held as first registered fingerprint image data Gr1.

  The second registered image memory 23c has a similarity equal to or greater than a threshold value even when a match determination between the first registered fingerprint image data Gr1 and the verified fingerprint image data Gs is obtained in accordance with the fingerprint verification process (see FIG. 7). The first registration fingerprint image Gr1 and the collation fingerprint image Gs slightly changed in fingerprint are additionally stored as second registration fingerprint image data Gr2 at the first time when it is determined that the similarity is not high.

  The collation image memory 23d stores fingerprint image data to be collated read by the image reading device 14 as collation fingerprint image data Gs.

  In the matching similarity memory 23e, a matching fingerprint image for the registered fingerprint image calculated in accordance with each fingerprint matching process (see FIG. 7) of the registered fingerprint image data (Gr1 / Gr2) and the matching fingerprint image data (Gs) is stored. Is stored.

  In the verification result memory 23f, the registered fingerprint image (Gr1 / Gr2 / Gr2) and the registered fingerprint image (Gr1 / Gr2) and the verification fingerprint image data (Gs) that are evaluated and determined in accordance with each fingerprint verification process (see FIG. 7) are stored. The determination result of match / non-match between Gr2) and the collation fingerprint image (Gs) is stored.

  In the fingerprint image collation apparatus of the second embodiment, the collation fingerprint image data Gs is determined to match the first registered fingerprint image data Gr1, and if the similarity is equal to or greater than a threshold, the collation fingerprint image Gs is updated. The first registered fingerprint image Gr1 is stored and updated, and if the similarity is less than the threshold value due to a temporary fingerprint change caused by environmental change or rough skin, the first registered fingerprint is stored. The verification fingerprint image Gs is added and stored as the second registered fingerprint image Gr2 without updating the storage of the image Gr1.

  Then, in a state in which the collation fingerprint image Gs that has undergone a temporary fingerprint change with respect to the first registered fingerprint image Gr1 is determined to be non-matching, the second registered fingerprint image Gr2 additionally registered due to the fingerprint change is used. If the collation similarity is equal to or greater than the threshold value, the second registered fingerprint image Gr2 is stored and updated in correspondence with the collation fingerprint image Gs.

  As a result, when the temporary fingerprint change is restored, a match determination can be obtained again based on the first registered fingerprint image Gr1.

  Next, the fingerprint registration / collation function of the second embodiment in the fingerprint image collation apparatus having the above configuration will be described in more detail.

(Fingerprint registration process)
FIG. 6 is a flowchart showing a fingerprint registration process for registering the fingerprint image of the legitimate user by the fingerprint image collation apparatus according to the second embodiment of the present invention.

  When the fingerprint image of a specific finger (for example, index finger) of the authorized user is read by the image reading device 14 by a preset number of image data and stored in the read image memory 23a in the RAM 13 (steps C1 and C2), For example, a read fingerprint image having the best imaging area and image quality is selected from the read fingerprint image data corresponding to the number of set data, and stored as first registered fingerprint image data Gr1 in the first registered image memory 23b ( Step C3).

(Fingerprint verification process)
FIG. 7 is a flowchart showing fingerprint collation processing by the fingerprint image collation apparatus according to the second embodiment of the present invention.

  When a fingerprint image of a specific finger (for example, forefinger) of the person to be verified is read by the image reading device 14 and stored in the read image memory 23a in the RAM 23, the read image is stored in the verification image memory 23d as verification fingerprint image data. It is stored as Gs (step D1).

  Then, the similarity between the first registered fingerprint image data Gr1 stored in the first registered image memory 23b and the collated fingerprint image data Gs stored in the collated image memory 23d is calculated, and collation determination processing is performed ( Step D2).

  Here, when the similarity between the first registered fingerprint image Gr1 and the collation fingerprint image Gs is equal to or higher than the predetermined similarity and collation matching is determined as the registrant's own fingerprint (step D3), the collation matching is further performed. It is determined whether or not the determined similarity is a high similarity equal to or higher than a preset threshold (step D4).

  Then, if it is determined that the similarity between the first registered fingerprint image Gr1 and the collation fingerprint image Gs that has been collated and matched is higher than a preset threshold value, collation fingerprint image data at that time Gs is stored and updated as the latest first registered fingerprint image data Gr1 (step D4 → D5).

  On the other hand, in Steps D1 to D3, although the collation fingerprint image Gs is determined to match the first registered fingerprint image Gr1 with a predetermined similarity or higher, the collation fingerprint image Gs has a period due to, for example, an environmental change or rough skin. If it is determined in step D4 that the collation similarity is not equal to or higher than a preset threshold value due to the occurrence of a slight and slight fingerprint change, the collation in which this slight and slight fingerprint change occurs The fingerprint image Gs is additionally stored as second registered fingerprint image data Gr2 in the second registered image memory 12c (steps D6 → D7).

  Thereafter, at the next fingerprint collation, in steps D1 to D3, the collation fingerprint image Gs is determined to match the first registered fingerprint image Gr1 with a predetermined similarity or more, but the collation fingerprint image Gs is temporarily received. If it is determined in step D4 that the collation similarity is not equal to or higher than a preset threshold because the target and slight fingerprint change continues to occur, the second registered image memory 23c is additionally registered. 2 The collation determination processing with the collation fingerprint image Gs is performed on the registered fingerprint image data Gr2 (steps D6 → D8). Then, the matching fingerprint image Gs is determined to match the second registered fingerprint image data Gr2 (step D9), and when it is determined that the similarity is a high similarity equal to or higher than a threshold (step D10), The collation fingerprint image data Gs at that time is stored and updated as the latest second registered fingerprint image data Gr2 whose fingerprint has changed temporarily (step D11).

  Further, at the time of the subsequent fingerprint collation, for example, because the registrant's temporary fingerprint change has progressed, the collation fingerprint image Gs is less than the predetermined similarity with respect to the first registered fingerprint image Gr1 in the steps D1 to D3. If a non-matching determination is made, a matching determination process with the matching fingerprint image Gs is performed on the second registered fingerprint image data Gr2 additionally updated and registered in the second registered image memory 23c (step D12). → D13). Even if the registrant's personal fingerprint changes for a while, the matching determination with the verification fingerprint image Gs can be obtained for the second registered fingerprint image data Gr2, thereby preventing erroneous determination due to the refusal of the user. (Step D14).

  Also in this case, when it is determined that the matching determination similarity of the verification fingerprint image Gs with respect to the second registered fingerprint image data Gr2 is a high similarity equal to or higher than a threshold value (step D15), the verification fingerprint at that time The image data Gs is stored and updated as the latest second registered fingerprint image data Gr2 whose fingerprint has changed temporarily (step D16).

  On the other hand, in the collation judgment processing (step D8 / D13) with the collation fingerprint image data Gs for the second registered fingerprint image data Gr2, of course, a match judgment is made when a non-coincidence judgment is made (step D9 / D14 (No)). Even if the match determination similarity is determined not to be a high similarity equal to or higher than a preset threshold value (step D10 / D15 (No)), the second registered fingerprint image data Gr2 is stored at that time. The update process for the verification fingerprint image data Gs is not performed (step D17).

  Therefore, according to the fingerprint registration / collation function of the second embodiment in the fingerprint image collation apparatus having the above configuration, the collation fingerprint image data Gs is determined to match the first registration fingerprint image data Gr1, and the similarity is equal to or greater than the threshold value. In some cases, the collation fingerprint image Gs is stored and updated as the latest first registered fingerprint image Gr1, and even if the matching is determined, the degree of similarity is due to a temporary fingerprint change due to environmental change or rough skin. If it is less than the threshold, the verification fingerprint image Gs is additionally stored as the second registered fingerprint image Gr2 without updating the storage of the first registered fingerprint image Gr1. Then, in a state in which the collation fingerprint image Gs that has undergone a temporary fingerprint change with respect to the first registered fingerprint image Gr1 is determined to be non-matching, the second registered fingerprint image Gr2 additionally registered due to the fingerprint change is used. The collation determination prevents the person from being rejected, and if the collation similarity is equal to or greater than the threshold, the second registered fingerprint image Gr2 is stored and updated in correspondence with the collation fingerprint image Gs.

  For this reason, it is possible not only to prevent misjudgment due to the refusal of the registrant when there is a temporary fingerprint change due to, for example, environmental changes or rough skin of the registrant, but also when the temporary fingerprint change is restored. In addition, the collation match determination can be obtained again based on the original first registered fingerprint image Gr1 without the refusal of the person.

  In each of the above-described embodiments, the case where the fingerprint image is the registration / collation target image has been described. However, for example, the partial image of each part of another living body such as an iris image or a face image is the registration / collation target image. However, the same effects as those of the above embodiments can be obtained.

  The processing method by the fingerprint image collation apparatus described in each of the above embodiments, that is, the fingerprint registration processing of the first embodiment for registering the fingerprint image of the authorized user shown in the flowchart of FIG. 3, and the flowchart of FIG. The fingerprint verification process of the first embodiment shown in FIG. 6, the fingerprint registration process of the second embodiment for registering the fingerprint image of the authorized user shown in the flowchart of FIG. 6, and the fingerprint verification process of the second embodiment shown in the flowchart of FIG. Each method such as memory card (ROM card, RAM card, etc.), magnetic disk (floppy disk, hard disk, etc.), optical disc (CD-ROM, DVD, etc.), semiconductor, etc. It can be stored and distributed in an external recording medium 18a such as a memory. Various computer terminals having a fingerprint reading function read the program stored in the external recording medium 18a into the storage device 12 by the storage medium reading unit 18, and the operation is controlled by the read program. The image registration / collation function described in each of the embodiments can be realized, and the same processing can be executed by the method described above.

  The program data for realizing each of the above methods can be transmitted on the communication network (Internet) 20 in the form of a program code, and a computer terminal (program server) connected to the communication network (Internet) 20 It is also possible to implement the image registration / collation function described above by fetching the program data from the storage device 22 of 21).

  Note that the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention at the stage of implementation. Further, 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 constituent requirements are deleted from all the constituent requirements shown in each embodiment or some constituent features are combined, the problems described in the column of the problem to be solved by the invention can be solved. When the effects described in the column of the effect of the invention can be obtained, a configuration in which these constituent elements are deleted or combined can be extracted as an invention.

The block diagram which shows the structure of the electronic circuit of the fingerprint image collation apparatus which concerns on embodiment of the image collation apparatus of this invention. The figure which shows the data memory ensured in RAM13 of the fingerprint image collation apparatus of 1st Embodiment of this invention. The flowchart which shows the fingerprint registration process for registering a regular user's own fingerprint image by the fingerprint image collation apparatus of 1st Embodiment of this invention. The flowchart which shows the fingerprint collation process by the fingerprint image collation apparatus of 1st Embodiment of this invention. The figure which shows the data memory ensured in RAM23 of the fingerprint image collation apparatus of 2nd Embodiment of this invention. The flowchart which shows the fingerprint registration process for registering a regular user's own fingerprint image by the fingerprint image collation apparatus of 2nd Embodiment of this invention. The flowchart which shows the fingerprint collation process by the fingerprint image collation apparatus of 2nd Embodiment of this invention.

Explanation of symbols

11 ... CPU
12 ... Storage device 13, 23 ... RAM
13a, 23a ... read image (G) memory 13b, 23b ... first registered image (Gr1) memory 13c ... first verification image (Gs1) memory 13d, 23c ... second registered image (Gs2) memory 13e ... second verification image (Gs2) memory 13f, 23e ... collation similarity memory 13g, 23f ... collation result memory 13h ... non-match count (n) memory 23d ... collation image (Gs) memory 14 ... image reading device 15 ... display unit 16 ... input unit 17 ... Bus 18 ... Storage medium reader 18a ... External storage medium 19 ... Transmission controller 20 ... Communication network (Internet)
21 ... Program server 22 ... Server storage device

Claims (8)

An image collation device that collates a registered image and a collation image to determine the identity of both,
A plurality of registered image storage means for storing images of a plurality of parts of the person to be registered;
Image reading means for reading a partial image of a person;
A first determination is made as to whether or not a registered image of a specific part of registered images of a plurality of parts stored by the plurality of registered image storage means matches with a collation image of the specific part read by the image reading means. Collation determination means;
When it is determined by the first collation determining means that the registered image of the specific part stored by the plural registered image storage means and the collated image of the specific part read by the image reading means are not matched, the plural A second collation determination unit that determines whether the registered image of the other part stored by the registered image storage unit and the collation image of the other part read by the image reading unit match or not;
When the first collation determining unit determines that the registered image of the specific part stored by the multiple registered image storage unit matches the collated image of the specific part read by the image reading unit, or the second When the matching determination unit determines that the registered image of the other part stored by the multiple registered image storage unit matches the verification image of the other part read by the image reading unit, the image reading A specific part registered image update storage unit that updates and stores a collation image of the specific part read by the unit as a registered image of the specific part stored by the plurality of registered image storage units;
An image collating apparatus comprising: further,
When it is determined by the second collation determination unit that the registered image of the other part stored by the plurality of registered image storage units matches the collation image of the other part read by the image reading unit, 2. The apparatus according to claim 1, further comprising another partial registered image update storage unit that updates and stores the collation image of the other part as a registered image of the other part stored by the plurality of registered image storage units. The image collation apparatus of description.   Matching or non-matching between the registered image and the matching image by the matching determination unit is performed when the registration image read by the image reading unit with respect to the registered image is less than a predetermined amount. The image collating apparatus according to claim 1, wherein the determination is based on a similarity with the collated image. An image collation device that collates a registered image and a collation image to determine the identity of both,
First registered image storage means for storing an image of a specific part of the person to be registered;
Image reading means for reading a partial image of a person;
First collation determining means for determining matching or non-coincidence based on the similarity between the registered image of the specific part stored by the first registered image storage means and the collation image of the specific part read by the image reading means When,
The first collation determination means determines the match between the registered image of the specific part stored in the first registered image storage means and the specific part of the collation image read by the image reading means, and the similarity of the match determination A first registered image update storage unit that updates and stores the collation image of the specific part as a registered image of the specific part stored by the first registered image storage unit when
The match between the registered image of the specific part stored by the first registered image storage unit and the collated image of the specific part read by the image reading unit is determined by the first verification determination unit, and the similarity of the match determination A second registered image storage means for storing the collation image of the specific part as a second registered image when
When it is determined by the first collation determining means that the registered image of the specific part stored by the first registered image storage means and the collation image of the specific part read by the image reading means are not matched, Second collation determining means for determining whether or not the second registered image of the specific part stored by the second registered image storage means matches the collation image of the specific part;
An image collating apparatus comprising: An image collation processing program for controlling a computer of an image collation apparatus that collates a registered image and a collation image to determine the identity of both,
The computer,
A plurality of registered image storage means for storing images of a plurality of parts of the person to be registered;
Image reading means for reading a partial image of a person,
A first determination is made as to whether or not a registered image of a specific part of registered images of a plurality of parts stored by the plurality of registered image storage means matches with a collation image of the specific part read by the image reading means. Collation determination means,
When it is determined by the first collation determining means that the registered image of the specific part stored by the plural registered image storage means and the collated image of the specific part read by the image reading means are not matched, the plural A second collation determining unit that determines whether the registered image of the other part stored by the registered image storage unit and the collated image of the other part read by the image reading unit match or not;
When the first collation determining unit determines that the registered image of the specific part stored by the multiple registered image storage unit matches the collated image of the specific part read by the image reading unit, or the second When the matching determination unit determines that the registered image of the other part stored by the multiple registered image storage unit matches the verification image of the other part read by the image reading unit, the image reading Specific part registered image update storage means for updating and storing a specific portion collation image read by the means as a registered image of the specific part stored by the plurality of registered image storage means;
A computer-readable image matching processing program designed to function as a computer. An image collation processing program for controlling a computer of an image collation apparatus that collates a registered image and a collation image to determine the identity of both,
The computer,
First registered image storage means for storing an image of a specific part of the person to be registered;
Image reading means for reading a partial image of a person,
First collation determining means for determining matching or non-coincidence based on a similarity between the registered image of the specific part stored by the first registered image storage means and the collation image of the specific part read by the image reading means. ,
The first collation determination means determines the match between the registered image of the specific part stored in the first registered image storage means and the specific part of the collation image read by the image reading means, and the similarity of the match determination A first registered image update storage unit that updates and stores the collation image of the specific part as a registered image of the specific part stored by the first registered image storage unit,
The match between the registered image of the specific part stored by the first registered image storage unit and the collated image of the specific part read by the image reading unit is determined by the first verification determination unit, and the similarity of the match determination Second registered image storage means for storing the specific portion of the collation image as the second registered image when the value is less than the predetermined value;
When it is determined by the first collation determination means that the registered image of the specific part stored by the first registered image storage means and the collation image of the specific part read by the image reading means are not matched, Second matching determination means for determining whether the second registered image of the specific part stored by the second registered image storage means matches or does not match the matching image of the specific part;
A computer-readable image matching processing program designed to function as a computer. An image collation method for collating a registered image and a collation image to determine the identity of both,
A plurality of registered image storage steps for storing images of a plurality of parts of the person to be registered;
An image reading step for reading a partial image of a person;
First, a determination is made as to whether or not the registered image of the specific portion of the registered images of the plurality of portions stored in the multiple registered image storage step matches the mismatched image of the specific portion read in the image reading step. 1 collation determination step;
In the first collation determination step, when it is determined that the registered image of the specific part stored in the multiple registered image storage step and the collation image of the specific part read in the image reading step are not matched, A second collation determining step for determining a match or non-coincidence between the registered image of the other part stored in the multiple registered image storing step and the collated image of the other part read in the image reading step; ,
When it is determined in the first collation determination step that the registered image of the specific part stored in the multiple registered image storage step matches the collation image of the specific part read by the image reading unit, or When it is determined in the second collation determination step that the registered image of the other part stored in the multiple registered image storage step matches the collation image of the other part read in the image reading step. A specific part registration image update storage step of updating and storing the collation image of the specific part read in the image reading step as a registration image of the specific part stored in the multiple registration image storage step;
An image collating method characterized by comprising: An image collation method for collating a registered image and a collation image to determine the identity of both,
A first registered image storing step for storing an image of a specific part of the person to be registered;
An image reading step for reading a partial image of a person;
A first collation for determining a match or non-coincidence based on the similarity between the registered image of the specific part stored in the first registered image storage step and the collation image of the specific part read in the image reading step A determination step;
In this first collation determination step, a match between the registered image of the specific portion stored in the first registered image storage step and the collation image of the specific portion read in the image reading step is determined, and the match determination A first registered image update storing step of updating and storing the collation image of the specific part as the registered image of the specific part stored in the first registered image storing step when the similarity is equal to or greater than a predetermined value;
In the first collation determining step, a match between the registered image of the specific portion stored in the first registered image storing step and the collated image of the specific portion read in the image reading step is determined, and the match determination A second registered image storing step of storing the collation image of the specific part as a second registered image when the similarity is less than a predetermined value;
When it is determined in the first collation determining step that the registered image of the specific portion stored in the first registered image storing step and the collated image of the specific portion read in the image reading step are not matched. A second collation determination step for determining whether the second registered image of the specific portion stored in the second registered image storage step matches or does not match the collation image of the specific portion;
An image collating method characterized by comprising:

Images