Patent Application
20060018515
This nonprovisional application is based on Japanese Patent Application No. 2004-197080 filed with the Japan Patent Office on Jul. 2, 2004, the entire contents of which are hereby incorporated by reference.
1. Field of the Invention
The present invention relates to a biometric data collating apparatus, a biometric data collating method and a biometric data collating program product, and particularly to a biometric data collating apparatus, a biometric data collating method and a biometric data collating program product which collates a collation target data formed of biometric information such as fingerprints with a plurality of collation data (i.e., data for collation).
2. Description of the Background Art
As a biometric data collating apparatus employing a biometrics technology, Japanese Patent Laying-Open No. 2003-323618 has disclosed such a biometric data collating apparatus that collates data of biometric information such as fingerprints provided thereto with collation data registered in advance for authenticating personal identification.
However, the conventional biometric data collating apparatus collates the collation target data provided thereto with the plurality of collation data by reading and using the collation data in an order fixed in advance, and cannot dynamically change the collation order for reducing a quantity or volume of processing. This results in problems that a processing quantity required for collation is large on average, and increases in proportion to the number of the registered collation data. Further, the large processing quantity results in a problem that the collation requires a long processing time and large power consumption.
An object of the invention is to reduce a processing quantity required for collating the input collation target data.
The above object of the invention can be achieved by a biometric data collating apparatus including the following components. Thus, the biometric data collating apparatus includes a collation target data input unit receiving biometric collation target data; a collation data storing unit storing a plurality of collation data used for collating the collation target data received by the collation target data input unit and an order of collation of the plurality of collation data; a collating unit reading each of the collation data stored in the collation data storing unit in the collation order, and collating the read collation data with the collation target data received by the collation target data input unit; and a collation order updating unit updating the collation order to put the collation data determined as matching data from the result of the collation by the collating unit in a leading place.
According to another aspect of the invention, a biometric data collating apparatus includes a collation target data input unit receiving biometric collation target data; a collation data storing unit storing a plurality of collation data used for collating the collation target data received by the collation target data input unit and priority values representing degrees of priority of collation for the respective collation data; a collating unit reading each of the collation data stored in the collation data storing unit in a descending order of the degree of the priority represented by the priority value, and collating the read collation data with the collation target data received by the collation target data input unit; and a priority value updating unit updating and changing the priority value corresponding to the collation data determined as matching data from the result of the collation by the collating unit into a value representing a higher degree of the priority.
According to still another aspect of the invention, a biometric data collating apparatus further includes a collation order updating unit updating the collation order of each of the collation data in the descending order of the degree of the priority represented by the priority value corresponding to the collation data. The collating unit reads the respective collation data in the collation order, and collates the read collation data with the collation target data received by the collation target data input unit. When the updated priority value corresponding to the collation data determined as the matching data from the result of the collation by the collating unit is larger than or equal to the priority value corresponding to the collation data preceding in the collation order the collation data determined as the matching data, the collation order updating unit replaces the places in the collation order of the above two collation data with each other.
The foregoing and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings.
Embodiments of the invention will now be described with reference to the drawings. A biometric information collating apparatus 1 receives biometric information data, and collates it with reference data (i.e., data for reference) which are registered in advance. Fingerprint image data will be described by way of example as collation target data, i.e., data to be collated. However, the data is not restricted to it, and may be another image data, voice data or the like representing another biometric feature which is similar to those of other individuals or persons, but never matches with them. Also, it may be image data of the striation or image data other than the striation. In the figures, the same or corresponding portions bear the same reference numbers, and description thereof is not repeated.
Referring to
The computer may be provided with a magnetic tape apparatus accessing to a cassette type magnetic tape that is detachably mounted thereto.
Referring to
Data input unit 101 includes a fingerprint sensor, and outputs a fingerprint image data that corresponds to the fingerprint read by the sensor. The sensor may be an optical, a pressure-type, a static capacitance type or any other type sensor.
Memory 102 includes a reference memory 1021 (i.e., memory for reference) storing data used for collation with the fingerprint image data applied to data input unit 101, a calculation memory 1022 temporarily calculating various calculation results, a taken-in data memory 1023 taking in the fingerprint image data applied to data input unit 101, and a collation order storing unit 1024 (i.e., memory for storing a collation order).
Collation processing unit 11 refers to each of the plurality of collation data (i.e., data for collation) stored in reference memory 1021, and determines whether the collation data matches with the fingerprint image data received by data input unit 101 or not. In the following description, the collation data stored in reference memory 1021 will be referred to as “reference data” hereinafter.
Collation order storing unit 1024 stores a collation order table including indexes of the reference data as elements. Biometric information collating apparatus 1 reads the reference data from reference memory 1021 in the order of storage in the collation order table, and collates them with the input fingerprint image data.
Bus 103 is used for transferring control signals and data signals between the units. Data correcting unit 104 performs correction (density correction) on data (i.e., fingerprint image in this embodiment) applied from data input unit 101. Maximum matching score position searching unit 105 uses a plurality of partial areas of one data (fingerprint image) as templates, and searches for a position of the other data (fingerprint image) that attains the highest matching score with respect to the templates. Namely, this unit serves as a so-called template matching unit.
Using the information of the result of processing by maximum matching score position searching unit 105 stored in memory 102, movement-vector-based similarity score calculating unit 106 calculates the movement-vector-based similarity score. Collation determining unit 107 determines a match/mismatch, based on the similarity score calculated by similarity score calculating unit 106. Control unit 108 controls processes performed by various units of collation processing unit 11.
Referring to
First, data input processing is executed (step T1). In the data input processing, control unit 108 transmits a data input start signal to data input unit 101, and thereafter waits for reception of a data input end signal. Data input unit 101 receiving the data input start signal takes in collation target data A for collation, and stores collation target data A at a prescribed address of taken-in data memory 1023 through bus 103. Further, after the input or take-in of collation target data A is completed, data input unit 101 transmits the data input end signal to control unit 108.
Then, the data correction processing is executed (step T2). In the data correction processing, control unit 108 transmits a data correction start signal to data correcting unit 104, and thereafter, waits for reception of a data correction end signal. In most cases, the input image has uneven image quality, as tones of pixels and overall density distribution vary because of variations in characteristics of data input unit 101, dryness of fingerprints and pressure with which fingers are pressed. Therefore, it is not appropriate to use the input image data directly for collation.
Data correcting unit 104 corrects the image quality of input image to suppress variations of conditions when the image is input (step T2). Specifically, for the overall image corresponding to the input image or small areas obtained by dividing the image, histogram planarization, as described in Computer GAZOU SHORI NYUMON (Introduction to computer image processing), SOKEN SHUPPAN, p. 98, or image thresholding (binarization), as described in Computer GAZOU SHORI NYUMON (Introduction to computer image processing), SOKEN SHUPPAN, pp. 66-69, is performed on collation target data A stored in taken-in data memory 1023. After the end of data correction processing of collation target data A, data correcting unit 104 transmits the data correction end signal to control unit 108.
Then, collation determining unit 107 performs collation determination on collation target data A subjected to the data correction processing by data correcting unit 104 and the reference data registered in advance in reference memory 1021 (step T3). The collation determination processing will be described later with reference to
Collation processing unit 11 performs the collation order updating processing (step T4). This processing updates the collation order table (see
Finally, control unit 108 outputs the result of the collation determination stored in memory 102 via display 610 or printer 690 (step T5). Thereby, the collation processing ends.
Referring to
Prior to the collation determination processing, control unit 108 transmits a collation determination start signal to collation determining unit 107, and waits for reception of a collation determination end signal.
In step S101, index ordidx of the element in the collation order table is initialized to 0 (first and thus Oth element).
In step S102, index ordidx of the element in the collation order table is compared with NREF, which is data representing the number of reference data stored in reference memory 1021. When index ordidx of the element in the collation order table is smaller than the number NREF of the reference data, the flow proceeds to step S103.
In step S103, Order[ordidx] is read from collation order storing unit 1024, and the read value is used as a value of a variable datidx.
In step S104, the reference data indicated by index datidx of the reference data is read from reference memory 1021, and the reference data thus read is used as data B.
In step S105, processing is performed to collate the input data (data A) with the read reference data (data B). This processing is formed of template matching and calculation of the similarity score. Procedures of this processing are illustrated in
First, control unit 108 transmits a template matching start signal to maximum matching score position searching unit 105, and waits for reception of a template matching end signal. Maximum matching score position searching unit 105 starts the template matching processing as illustrated in steps S001 to S007. In step S001, a variable i of a counter is initialized to 1. In step S002, an image of a partial area, which is defined as a partial region Ri, is set as a template to be used for the template matching.
Though the partial area Ri has a rectangular shape for simplicity of calculation, the shape is not limited thereto. In step S003, processing is performed to search for a position, where data B exhibits the highest matching score with respect to the template set in step S002, i.e., the position where matching of data in the image is achieved to the highest extent. More specifically, it is assumed that partial area Ri used as the template has an image density of Ri(x, y) at coordinates (x, y) defined based on its upper left corner, and data B has an image density of B(s, t) at coordinates (s, t) defined based on its upper left corner. Also, partial area Ri has a width w and a height h, and each of pixels of data A and B has a possible maximum density of V0. In this case, a matching score Ci(s, t) at coordinates (s, t) of data B can be calculated based on density differences of respective pixels according to the following equation (1).
In data B, coordinates (s, t) are successively updated and matching score C(s, t) in coordinates (s, t) is calculated. A position having the highest value is considered as the maximum matching score position, the image of the partial area at that position is represented as partial area Mi, and the matching score at that position is represented as maximum matching score Cimax. In step S004, maximum matching score Cimax in data B for partial area Ri calculated in step S003 is stored at a prescribed address of memory 1022. In step S005, a movement vector Vi is calculated in accordance with the following equation (2), and is stored at a prescribed address of memory 1022.
As already described, processing is effected based on partial area Ri corresponding to position P set in data A, and data B is scanned to determine a partial area Mi in a position M exhibiting the highest matching score with respect to partial area Ri. A vector from position P to position M thus determined is referred to as the “movement vector”. This is because data B seems to have moved from data A as a reference, as the finger is placed in various manners on the fingerprint sensor.
Vi=(Vix,Viy)=(Mix−Rix,Miy−Riy) (2)
In the above equation (2), variables Rix and Riy are x and y coordinates of the reference position of partial area Ri, and correspond, by way of example, to the upper left corner of partial area Ri in data A. Variables Mix and Miy are x and y coordinates of the position of maximum matching score Cimax, which is the result of search of partial area Mi, and correspond, by way of example, to the upper left corner coordinates of partial area Mi located at the matched position in data B.
In step S006, it is determined whether counter variable i is smaller than a maximum value n of the index of the partial area or not. If the value of variable i is smaller than n, the process proceeds to step S007, and otherwise, the process proceeds to step S008. In step S007, 1 is added to the value of variable i. Thereafter, as long as the value of variable i is not larger than n, steps S002 to S007 are repeated. By repeating these steps, template matching is performed for each partial area Ri to calculate maximum matching score Cimax and movement vector Vi of each partial area Ri.
Maximum matching score position searching unit 105 stores maximum matching score Cimax and movement vector Vi for every partial area Ri, which are calculated successively as described above, at prescribed addresses, and thereafter transmits the template matching end signal to control unit 108. Thereby, the process proceeds to step S008.
Thereafter, control unit 108 transmits a similarity score calculation start signal to movement-vector-based similarity score calculating unit 106, and waits for reception of a similarity score calculation end signal. Movement-vector-based similarity score calculating unit 106 calculates the similarity score through the process of steps S008 to S020 of
In step S008, similarity score P(A, B) is initialized to 0. Here, similarity score P(A, B) is a variable storing the degree of similarity between data A and B. In step S009, index i of movement vector Vi used as a reference is initialized to 1. In step S010, similarity score Pi related to movement vector Vi used as the reference is initialized to 0. In step S011, index j of movement vector Vj is initialized to 1. In step S012, a vector difference dVij between reference movement vector Vi and movement vector Vj is calculated in accordance with the following equation (3).
dVij=|Vi−Vj|=sqrt((Vix−Vjx)ˆ2+(Viy−Vjy){circumflex over (2)}) (3)
Here, variables Vix and Viy represent components in x and y directions of movement vector Vi, respectively, and variables Vjx and Vjy represent components in x and y directions of movement vector Vj, respectively. Variable sqrt(X) represents a square root of X, and Xˆ2 is an equation calculating a square of X.
In step S013, vector difference dVij between movement vectors Vi and Vj is compared with a prescribed constant ε, and it is determined whether movement vectors Vi and Vj can be regarded as substantially the same vectors or not. If vector difference dVij is smaller than the constant ε, movement vectors Vi and Vj are regarded as substantially the same, and the flow proceeds to step S014. If the difference is larger than the constant, the movement vectors cannot be regarded as substantially the same, and the flow proceeds to step S015. In step S014, similarity score Pi is incremented in accordance with the following equations (4) to (6).
Pi=Pi+α (4)
α=1 (5)
α=Cjmax (6)
In equation (4), variable α is a value for incrementing similarity score Pi. If cc is set to 1 as represented by equation (5), similarity score Pi represents the number of partial areas that have the same movement vector as reference movement vector Vi. If α is equal to Cjmax as represented by equation (6), similarity score Pi is equal to the total sum of the maximum matching scores obtained through the template matching of partial areas that have the same movement vectors as the reference movement vector Vi. The value of variable α may be reduced depending on the magnitude of vector difference dVij.
In step S015, it is determined whether index j is smaller than the value n or not. If index j is smaller than n, the flow proceeds to step S016. Otherwise, the flow proceeds to step S017. In step S016, the value of index j is incremented by 1. By the process from step S010 to S016, similarity score Pi is calculated, using the information of partial areas determined to have the same movement vector as the reference movement vector Vi. In step S017, similarity score Pi using movement vector Vi as a reference is compared with variable P(A, B). If similarity score Pi is larger than the largest similarity score (value of variable P(A, B)) obtained by that time, the flow proceeds to step S018, and otherwise the flow proceeds to step S019.
In step S018, variable P(A, B) is set to a value of similarity score Pi using movement vector Vi as a reference. In steps S017 and S018, if similarity score Pi using movement vector Vi as a reference is larger than the maximum value of the similarity score (value of variable P(A, B)) calculated by that time using another movement vector as a reference, the reference movement vector Vi is considered to be the best reference among movement vectors Vi, which have been represented by index i.
In step S019, the value of index i of reference movement vector Vi is compared with the maximum value (value of variable n) of the indexes of partial areas. If index i is smaller than the number of partial areas, the flow proceeds to step S020, in which index i is incremented by 1. Otherwise, the flow in
By the processing from step S008 to step S020, similarity between image data A and B is calculated as the value of variable P(A, B). Movement-vector-based similarity score calculating unit 106 stores the value of variable P(A, B) calculated in the above described manner at a prescribed address of memory 1022, and transmits a similarity score calculation end signal to control unit 108 to end the process.
Referring to
When it is determined in step S102 that updated ordidx is not smaller than number NREF of the reference data, this means that there is no reference data matching with input data A. In this case, a value, e.g., of “−1” representing “mismatching” is written into a prescribed address of calculation memory 1022 (step S109). Further, the collation determination end signal is transmitted to control unit 108, and the process ends.
First, in step U101, a result of collation, which is written in step S108 or S109, is read from calculation memory 1022, and it is determined whether the result of collation represents “mismatching” or not. If it represents “mismatching”, a collation order updating end signal is transmitted to control unit 108 to end the processing. If it is determined in step U101 that the result represents “matching”, the flow proceeds to step U102.
In step U102, the value of variable j is initialized to index ordidx which is attained in the collation order table at the time of the matching of the reference data. In other words, the value of variable j is updated to the value of ordidx which is written as the collation result into the prescribed address of calculation memory 1022 in step S108.
For example, when calculation memory 1022 has stored the collation results representing that the collation target data matches with reference data C in
In step U103, the value of variable j is compared with 0. If j is larger than 0, processing from step U103 to step U105 is performed. When j becomes equal to 0, processing in step U106 is performed. For example, if variable j is “2”, the flow proceeds to step U104.
In step U104, the value of Order[j−1] is written into Order[j]. For example, when j is “2” in the collation order table of
In step U105, 1 is subtracted from the value of j, and the processing starting from step U103 is repeated. Consequently, in the collation order table, e.g., of
In step U106, index datidx of the reference data at the time of matching is written into Order[0]. Thereby, the matching reference data becomes a first element in the collation order data. For example, in the collation order table of
A second embodiment will now be described with reference to
In response to every determination as “matching” in the collation determination, collation processing unit 11 adds a predetermined value (e.g., of “1”) to the collation frequency value of the reference data determines as “matching”. Therefore, a larger collation frequency value represents a higher collation frequency. In this second embodiment, the reference order of the reference data is updated in the descending order of the collation frequency.
The procedures of the collation processing executed by biometric information collating apparatus 1 of the second embodiment are substantially the same as those of the collation processing of the first embodiment. Therefore, biometric information collating apparatus 1 of the second embodiment executes the processing illustrated in
Referring to
In step U201, the result of the collation, which was written in step S108 or S109, is read from calculation memory 1022, and it is determined whether the collation result is “mismatching” or not. If “mismatching”, the collation order updating end signal is transmitted to control unit 108, and the processing ends. If it is determined in step U201 that the collation result is “matching”, the flow proceeds to step U202.
In step U202, a predetermined updating value is added to collation frequency value Freq[ordidx] corresponding to index ordidx in the collation order table at the time of matching of the reference data. In connection with this, ordidx is a value which is written as the collation result into a prescribed address of calculation memory 1022 in step S108. The updating value is, e.g., “1”.
When calculation memory 1022 has stored the collation result representing the matching of the collation target data with reference data C in
The updating value is not restricted to “1”. Normalization may be performed such that a sum of all the collation frequency values in the collation frequency table may take a constant value, and the collation frequency value may be a stochastic value.
In step U203, the value of variable j is initialized to index ordidx in the collation order table appearing at the time of matching of the reference data. In other words, the value of variable j is updated to the value of ordidx which is written as a collation result into the prescribed address of calculation memory 1022 in step S108.
For example, when calculation memory 1022 has stored the collation result representing the matching of the collation target data with reference data C in
In step U204, the value of variable j is compared with 0. While j is larger than 0, the flow proceeds to step U205. When j matches with 0, the collation order updating end signal is transmitted to control unit 108, and the processing ends. For example, when variable j is “2”, the flow proceeds to step U205.
In step U205, the value of Freq[j−1] is compared with the value of Freq[j]. If the former is larger than the latter, the collation order updating end signal is transmitted to control unit 108. Otherwise, processing in step U206 is performed.
For example, when a comparison is made between the values of Freq[2−1] and Freq[2] in
In step U206, the values of Order[j−1] and Order[j] are replaced with each other in the collation order table. Order[j] means the reference data in the collation order table corresponding to index j. In subsequent step U207, the values of Freq[j−1] and Freq[j] are replaced with each other in the collation frequency table.
For example, the values of Order[2−1] and Order[2] are replaced with each other in
In step U208, 1 is subtracted from the value of j, and the processing in and after step U204 is repeated. Consequently, in the updated collation order table, e.g., in
According to the second embodiment, the reference order of the reference data is updated in the descending order of the frequency of matching as a result of the collation determination. Therefore, the collation determination can be performed by successively referring to the reference data in the descending order of the probability of matching. Consequently, the time of the collation processing can be reduced on average.
The processing function for collation already described are achieved by programs. According to a third embodiment, such programs are stored on computer-readable recording medium.
In the third embodiment, the recording medium may be a memory required for processing by the computer show in
The above recording medium can be separated from the computer body. A medium stationarily bearing the program may be used as such recording medium. More specifically, it is possible to employ tape mediums such as a magnetic tape and a cassette tape as well as disk mediums including magnetic disks such as FD 632 and fixed disk 626, and optical disks such as CD-ROM 642, MO (Magnetic Optical) disk, MD (Mini Disk) and DVD (Digital Versatile Disk), card mediums such as an IC card (including a memory card) and optical card, and semiconductor memories such as a mask ROM, EPROM (Erasable and Programmable ROM), EEPROM (Electrically EPROM) and flash ROM.
Since the computer in
The form of the contents stored on the recording medium is not restricted to the program, and may be data.
According to the invention relating to the embodiments already described, the order of the collation of the reference data with the input collation target data is dynamically changed so that it can be expected to reduce the quantity of processing of the data collation. This effect is particularly effective in the case where the reference data is used in an unbalanced fashion. The precise biometric information collation, which is less sensitive to presence/absence of minutiae, number and clearness of images, environmental change at the time of image input, noises and others, can be performed in a short collation time with reducible power consumption. The reduction of processing is automatically performed, and this effect can be maintained without requiring the maintenance of the device.
Although the present invention has been described and illustrated in detail, it is clearly understood that the same is by way of illustration and example only and is not to be taken by way of limitation, the spirit and scope of the present invention being limited only by the terms of the appended claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2004-197080(P) | Jul 2004 | JP | national |
