JP2000348178A - Fingerprint collation device - Google Patents
Fingerprint collation deviceInfo
- Publication number
- JP2000348178A JP2000348178A JP11162290A JP16229099A JP2000348178A JP 2000348178 A JP2000348178 A JP 2000348178A JP 11162290 A JP11162290 A JP 11162290A JP 16229099 A JP16229099 A JP 16229099A JP 2000348178 A JP2000348178 A JP 2000348178A
- Authority
- JP
- Japan
- Prior art keywords
- collation
- fingerprint
- fingerprint image
- center position
- straight line
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1365—Matching; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Collating Specific Patterns (AREA)
- Image Analysis (AREA)
Abstract
Description
【０００１】[0001]
【発明の属する技術分野】本発明は、指紋照合装置に係
り、特に、指紋画像の画像データに基づいて指紋照合を
行う指紋照合装置に関する。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fingerprint matching device, and more particularly, to a fingerprint matching device that performs fingerprint matching based on image data of a fingerprint image.
【０００２】[0002]
【従来の技術】近年、指紋を照合することにより個人を
認識する指紋照合装置が提案されている。例えば、特開
平９−２２４０６号公報に記載された装置では、予め登
録した指紋の画像データに二次元離散的フーリエ変換を
施し、フーリエ画像データに変換する。そして、照合す
る指紋の画像データにも二次元離散的フーリエ変換を施
してフーリエ画像データに変換し、登録指紋のフーリエ
画像データと合成する。2. Description of the Related Art In recent years, a fingerprint collation apparatus for recognizing an individual by collating a fingerprint has been proposed. For example, in an apparatus described in Japanese Patent Application Laid-Open No. 9-22406, two-dimensional discrete Fourier transform is performed on fingerprint image data registered in advance to convert the data into Fourier image data. Then, the image data of the fingerprint to be collated is also subjected to a two-dimensional discrete Fourier transform to be converted into Fourier image data, and combined with the Fourier image data of the registered fingerprint.
【０００３】次に、合成したフーリエ画像データに対し
て振幅抑制処理であるｌｏｇ処理を行い、振幅抑制処理
された合成フーリエ画像データに対して更に二次元的離
散フーリエ変換を施す。そして、相関成分エリアの相関
成分の強度から相関値を求め、この相関値が所定しきい
値以上の場合には登録指紋と照合指紋とが一致したと判
定する。Next, log processing as amplitude suppression processing is performed on the synthesized Fourier image data, and two-dimensional discrete Fourier transform is further performed on the synthesized Fourier image data subjected to the amplitude suppression processing. Then, a correlation value is obtained from the intensity of the correlation component in the correlation component area. If the correlation value is equal to or greater than a predetermined threshold value, it is determined that the registered fingerprint and the verification fingerprint match.
【０００４】[0004]
【発明が解決しようとする課題】しかしながら、上記従
来技術では、照合指紋の画像データに対する二次元離散
的フーリエ変換、登録指紋のフーリエ画像データとの照
合、振幅抑制処理、及び相関値を求めるための二次元離
散的フーリエ変換の各演算処理が、画像データについて
画素単位に繰り返し行われるため、処理時間が長くなる
という問題があった。However, in the above-mentioned prior art, two-dimensional discrete Fourier transform of the image data of the collation fingerprint, collation with the Fourier image data of the registered fingerprint, amplitude suppression processing, and calculation of the correlation value are performed. Since each operation process of the two-dimensional discrete Fourier transform is repeatedly performed on the image data in pixel units, there is a problem that the processing time is long.
【０００５】本発明は、上記問題を解決すべく成された
ものであり、指紋照合の処理時間を大幅に短縮すること
ができる指紋照合装置を提供することが目的である。SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problem, and has as its object to provide a fingerprint matching device capable of greatly reducing the processing time of fingerprint matching.
【０００６】[0006]
【課題を解決するための手段】上記目的を達成するため
に、請求項１記載の発明の指紋照合装置は、照合指紋画
像の紋様の中心位置を検出する中心位置検出手段と、検
出した中心位置を通る予め定めた複数の直線上の画素の
明暗信号を各々検出する明暗信号検出手段と、検出した
複数の明暗信号を予め定めた指紋の隆線の平均ピッチに
相当する周波数よりも高い周波数で時間・周波数領域信
号へ変換する変換手段と、前記複数の直線の時間・周波
数領域信号の所定成分と、予め定めた登録指紋画像につ
いて該画像の紋様の中心位置を通り、前記予め定めた複
数の直線に相当する直線上の画素列に関する時間・周波
数領域信号の所定成分とに基づいて各直線の相互相関値
を算出し、該算出した相互相関値に基づいて各直線の一
致度を各々算出する一致度算出手段と、各々算出された
一致度に基づいて前記照合指紋画像と前記登録指紋画像
との照合を行う照合手段と、を有することを特徴として
いる。According to a first aspect of the present invention, there is provided a fingerprint matching apparatus, comprising: a center position detecting means for detecting a center position of a pattern of a matching fingerprint image; Light and dark signal detecting means for respectively detecting light and dark signals of pixels on a predetermined plurality of straight lines passing through, a plurality of detected light and dark signals at a frequency higher than the frequency corresponding to the average pitch of the predetermined fingerprint ridges A conversion unit for converting to a time / frequency domain signal, a predetermined component of the plurality of linear time / frequency domain signals, and a predetermined registered fingerprint image passing through the center position of the pattern of the image, and A cross-correlation value of each straight line is calculated based on a predetermined component of a time / frequency domain signal related to a pixel row on a straight line corresponding to the straight line, and a degree of coincidence of each straight line is calculated based on the calculated cross-correlation value. And matching degree calculation means, is characterized by having a matching unit for matching between the registered fingerprint image and the collation fingerprint image on the basis of each calculated degree of matching, a.
【０００７】請求項１記載の発明によれば、中心位置検
出手段は、照合指紋画像の紋様の中心位置を検出する。
この中心位置は、例えば指紋が楕円状の場合には、この
楕円の長軸と短軸の交点を紋様の中心位置として検出す
ることができる。また、照合指紋画像にエッジ強調等公
知の画像処理を施し、この画像処理後の照合指紋画像か
ら中心位置を検出するようにしてもよい。According to the first aspect of the present invention, the center position detecting means detects the center position of the pattern of the collation fingerprint image.
For example, when the fingerprint is elliptical, the center position can be detected as the center position of the pattern at the intersection of the major axis and the minor axis of the ellipse. Further, a known image process such as edge enhancement may be applied to the collation fingerprint image, and the center position may be detected from the collation fingerprint image after the image processing.
【０００８】明暗信号検出手段は、画素の濃度等を検出
することにより、中心位置検出手段により検出された中
心位置を通る複数の直線を予め定め、この複数の直線上
の画素の明暗信号を各々検出する。複数の直線は、例え
ば予め基準となる直線を定め、この基準となる直線を、
中心位置を原点として等角度ずつ回転させた直線として
定める。また、等角度ずつ回転させず、異なる角度で回
転させた直線として定めてもよい。The light / dark signal detecting means determines a plurality of straight lines passing through the center position detected by the center position detecting means by detecting the density of the pixels and the like, and outputs the light / dark signals of the pixels on the plurality of straight lines respectively. To detect. For the plurality of straight lines, for example, a reference straight line is determined in advance, and this reference straight line is
The center position is defined as a straight line rotated by equal angles from the origin. Alternatively, the rotation may be determined as a straight line rotated at a different angle without being rotated at an equal angle.
【０００９】変換手段は、各々検出した複数の明暗信号
を時間・周波数領域信号へ変換する。この変換は、請求
項４にも記載したように、時間分解能に優れたウェーブ
レット変換により行うことができる。このとき、変換周
波数は、予め定めた指紋の隆線の平均ピッチに相当する
周波数よりも高い周波数とする。これにより、指紋の隆
線の明暗をより強調することができる。The conversion means converts each of the detected light and dark signals into a time / frequency domain signal. This conversion can be performed by a wavelet transform having excellent time resolution, as described in claim 4. At this time, the conversion frequency is higher than a frequency corresponding to the predetermined average pitch of the ridges of the fingerprint. Thereby, the brightness of the ridge of the fingerprint can be further emphasized.
【００１０】一致度算出手段は、照合指紋画像の紋様の
中心位置を通る複数の直線の時間・周波数領域信号の所
定成分と、予め定めた登録指紋画像の紋様の中心位置を
通る予め定めた複数の直線上の画素の明暗信号を変換し
て得られた各々の時間・周波数領域信号の所定成分とに
基づいて各直線の相互相関値を算出する。なお、所定成
分は、請求項３にも記載したように、パワー成分及び位
相成分の少なくとも一方とすることができる。パワー成
分及び位相成分の両方に基づいて相互相関値を算出すれ
ば、一致度の精度を高めることができる。The coincidence calculating means includes a predetermined component of the time / frequency domain signal of a plurality of straight lines passing through the center position of the pattern of the verification fingerprint image and a predetermined plurality of components passing through the center position of the predetermined registered fingerprint image. The cross-correlation value of each straight line is calculated based on the predetermined components of each time / frequency domain signal obtained by converting the light and dark signals of the pixels on the straight line. Note that the predetermined component can be at least one of a power component and a phase component. If the cross-correlation value is calculated based on both the power component and the phase component, the accuracy of the degree of coincidence can be improved.
【００１１】照合手段は、各々算出された一致度に基づ
いて前記照合指紋画像と前記登録指紋画像との照合を行
う。例えば、各々算出された一致度のうち、所定値以上
のものが所定個以上ある場合を指紋が一致したと判断す
ることができる。このように、２次元の指紋画像を指紋
画像の中心位置を通る複数の直線の明暗信号、すなわち
１次元の信号として検出し、この検出した１次元の明暗
信号を時間・周波数領域へ変換して処理するので、すべ
ての画素について変換処理する場合と比較して処理時間
を大幅に短縮することができると共に、照合指紋画像の
位置ずれに影響されることなく、正確に照合することが
できる。The collating means performs collation between the collated fingerprint image and the registered fingerprint image based on the calculated degree of coincidence. For example, among the calculated degrees of coincidence, a case where the number of fingerprints equal to or more than a predetermined value is equal to or more than a predetermined value can be determined as a fingerprint match. As described above, the two-dimensional fingerprint image is detected as a plurality of straight light / dark signals passing through the center position of the fingerprint image, that is, a one-dimensional signal, and the detected one-dimensional light / dark signal is converted into a time / frequency domain. Since the processing is performed, the processing time can be significantly reduced as compared with the case where the conversion processing is performed on all the pixels, and accurate collation can be performed without being affected by the displacement of the collation fingerprint image.
【００１２】また、一致度算出手段は、請求項２にも記
載したように、照合指紋画像の中心位置を通る複数の直
線の中の予め定めた直線に関する時間・周波数領域信号
の所定成分と、登録指紋画像の中心位置を通る複数の直
線に関する時間・周波数領域信号の所定成分との相互相
関値を各々算出し、該算出した相互相関値が最も高い直
線を基準としてその他の各直線の一致度を各々算出する
ことができる。このように、照合指紋画像の中心位置を
通る複数の直線の中から１つの直線を予め定め、この予
め定めた直線に関する時間・周波数領域信号の所定成分
と登録指紋画像の複数の直線に関する時間・周波数領域
信号の所定成分との相互相関値が最も高い直線を基準と
して一致度を算出するので、照合指紋画像と登録指紋画
像とに回転ずれが有る場合でも正確に照合することがで
きる。[0012] In addition, the coincidence calculating means includes a predetermined component of a time / frequency domain signal related to a predetermined straight line among a plurality of straight lines passing through the center position of the collation fingerprint image. Calculate the cross-correlation values of a plurality of straight lines passing through the center position of the registered fingerprint image with a predetermined component of the time / frequency domain signal, and determine the degree of coincidence of each of the other straight lines based on the straight line having the highest calculated cross-correlation value Can be calculated respectively. As described above, one straight line is determined in advance from a plurality of straight lines passing through the center position of the collation fingerprint image, and a predetermined component of the time / frequency domain signal related to the predetermined straight line and the time / time related to the plurality of straight lines of the registered fingerprint image are determined. Since the degree of coincidence is calculated based on the straight line having the highest cross-correlation value with the predetermined component of the frequency domain signal, accurate collation can be performed even when there is a rotational displacement between the collation fingerprint image and the registered fingerprint image.
【００１３】[0013]
【発明の実施の形態】以下、図面を参照して本発明の実
施の形態を詳細に説明する。Embodiments of the present invention will be described below in detail with reference to the drawings.
【００１４】図１には、本実施の形態に係る指紋照合装
置１０が示されている。指紋照合装置１０は、中心位置
検出部１２、明暗信号検出部１４、ウェーブレット変換
部１６、照合部１８、及びメモリ２０で構成されてい
る。FIG. 1 shows a fingerprint matching apparatus 10 according to the present embodiment. The fingerprint matching device 10 includes a center position detecting unit 12, a light / dark signal detecting unit 14, a wavelet transform unit 16, a matching unit 18, and a memory 20.
【００１５】中心位置検出部１２は、図示しないＣＣＤ
カメラ等により撮像した照合すべき指紋画像（以下、照
合指紋画像という）の画像データに対してエッジ強調等
の従来より公知の種々の画像処理方法を施すことにより
求められた個人を特定する隆線（指紋の凸部）画像を取
りこむ。そして、この隆線画像の紋様の中心位置を検出
する。The center position detector 12 is a CCD (not shown)
A ridge that specifies an individual obtained by performing various conventionally known image processing methods such as edge enhancement on image data of a fingerprint image to be verified (hereinafter referred to as a verification fingerprint image) captured by a camera or the like. (Fingerprint of fingerprint) Capture an image. Then, the center position of the pattern of the ridge image is detected.
【００１６】例えば、求めた隆線画像が、図２（Ａ）に
示すような楕円状の隆線画像の場合には、まず楕円を特
定し、特定した楕円の長軸と短軸との交点を求め、これ
を紋様の中心位置Ｘとして検出することができる。ま
た、図２（Ｂ）に示すように、紋様が１つの渦巻き状の
渦巻き線になっている場合には、この渦巻き線の始点
（渦巻き線の中心部側の端点）を紋様の中心位置Ｘとし
て検出することができる。また、図２（Ｃ）に示すよう
に、紋様が複数の渦巻き状の渦巻き線によりなっている
場合には、それぞれの渦巻き線の始点の中間位置を検出
し、これを紋様の中心位置Ｘとすることができる。ま
た、図２（Ｄ）に示すように、隆線の形状が楕円状にな
っていない場合には、紋様の中心にある隆線の変曲点を
検出し、これを中心位置Ｘとすることができる。この変
曲点は、例えば紋様の最も内側にある隆線の接線を微分
し、この微分値が０となる接線と隆線との交点を検出す
ることにより検出することができる。For example, if the obtained ridge image is an elliptical ridge image as shown in FIG. 2A, the ellipse is first specified, and the intersection of the specified long and short axes of the ellipse is determined. And this can be detected as the center position X of the pattern. In addition, as shown in FIG. 2B, when the pattern is a single spiral line, the starting point of this spiral line (the end point on the center side of the spiral line) is the center position X of the pattern. Can be detected as Further, as shown in FIG. 2C, when the pattern is composed of a plurality of spiral lines, an intermediate position between the starting points of the spiral lines is detected, and this is defined as the central position X of the pattern. can do. In addition, as shown in FIG. 2D, when the shape of the ridge is not elliptical, the inflection point of the ridge at the center of the pattern is detected and set as the center position X. Can be. This inflection point can be detected, for example, by differentiating the tangent of the innermost ridge of the pattern and detecting the intersection of the tangent and the ridge whose differential value is zero.
【００１７】明暗信号検出部１４は、図３に示すよう
に、例えば中心位置Ｘを通る図３において水平な直線Ａ
1−ａ1を基準として定め、これを時計回りに１８０度の
範囲で等角度ずつ回転して得られるＮ個の直線を定め
る。例えば１０度ずつ回転させた場合には、１８本の直
線が定められる。そして、このＮ本の直線Ａ1−ａ1、Ａ
2−ａ2、・・・Ａn−ａn上の画素列の濃度を検出するこ
とにより明暗信号を各々の直線について検出する。な
お、基準となる直線は任意に設定することができる。ま
た、複数の直線のパターンを予め定めておき、このパタ
ーンを照合指紋画像に当てはめるようにしてもよい。As shown in FIG. 3, the light / dark signal detector 14 is, for example, a horizontal straight line A in FIG.
Defining a 1 -a 1 as a reference, determine N number of straight line obtained which was rotated at equal angles in a range of 180 degrees clockwise. For example, when rotated by 10 degrees, 18 straight lines are defined. Then, the N straight lines A 1 -a 1 , A
2 -a 2, detects light and dark signals for each of the straight line by detecting the density of a pixel row on ··· A n -a n. The reference straight line can be set arbitrarily. Alternatively, a plurality of straight line patterns may be determined in advance, and this pattern may be applied to the collation fingerprint image.
【００１８】明暗信号検出部１４において検出した直線
Ａ1−ａ1の明暗信号の例を図４（Ａ）に示す。図４
（Ａ）の縦軸は、明暗の度合いを示し、横軸は、画像位
置を時間に変換したものを示している。ここで、明部
は、指紋の谷線部（凹部）を示し、暗部は、指紋の隆線
部（凸部）を示している。すなわち、この明暗信号は、
指紋の隆線（又は谷線）のピッチ（間隔）に対応してい
る。FIG. 4A shows an example of a light / dark signal of a straight line A 1 -a 1 detected by the light / dark signal detecting section 14. FIG.
The vertical axis of (A) indicates the degree of lightness and darkness, and the horizontal axis indicates the image position converted to time. Here, a bright portion indicates a valley portion (concave portion) of the fingerprint, and a dark portion indicates a ridge portion (convex portion) of the fingerprint. That is, this light / dark signal is
It corresponds to the pitch (interval) of the ridges (or valleys) of the fingerprint.
【００１９】ウェーブレット変換部１６は、図４（Ａ）
に示された明暗信号を時間・周波数領域信号へ変換す
る。この時間・周波数領域信号への変換は、時間分解能
で優れた公知のウェーブレット変換を用いて行う。な
お、ウェーブレット変換周波数は、前記隆線の平均ピッ
チをＴとした場合の周波数ｆ（＝１／Ｔ）の所定倍（例
えば２倍）以上の周波数とする。これにより、明暗の変
化、すなわち隆線部を強調することができる。図４
（Ｂ）に、図４（Ａ）に示す明暗信号に対してウェーブ
レット変換を施して得られた時間・周波数領域信号のパ
ワー成分を示す。このように、指紋の谷線部はパワーが
小さくなり、指紋の隆線部はパワーが大きくなる。ウェ
ーブレット変換部１６では、このようなウェーブレット
変換を残りの直線についてそれぞれ行う。FIG. 4A shows the wavelet transform section 16.
Is converted into a time / frequency domain signal. The conversion into the time / frequency domain signal is performed using a known wavelet transform having excellent time resolution. Note that the wavelet transform frequency is a frequency that is equal to or more than a predetermined multiple (for example, twice) of the frequency f (= 1 / T) when the average pitch of the ridges is T. This makes it possible to emphasize a change in brightness, that is, a ridge portion. FIG.
FIG. 4B shows a power component of a time / frequency domain signal obtained by performing a wavelet transform on the light / dark signal shown in FIG. As described above, the valley portion of the fingerprint has a lower power, and the ridge portion of the fingerprint has a higher power. The wavelet transform unit 16 performs such a wavelet transform on each of the remaining straight lines.
【００２０】メモリ２０には、照合指紋画像の照合対称
となる指紋画像（以下、登録指紋画像という）につい
て、前述した登録指紋画像の中心位置の検出、検出した
中心位置を通る複数の直線の明暗信号の検出、検出した
明暗信号の時間・周波数領域信号への変換の各処理を予
め行うことにより得られた登録指紋画像の直線Ｂ1−ｂ1
〜Ｂn−ｂnの時間・周波数領域信号のパワー成分値が記
憶されている。この登録指紋画像の直線Ｂ1−ｂ1〜Ｂn
−ｂnは、照合指紋画像の直線Ａ1−ａ1〜Ａn−ａnと同
様に、水平な直線Ｂ1−ｂ1を基準として、これを時計回
りに１８０度の範囲で等角度ずつ回転して得られるＮ個
の直線である。なお、隣接する各直線が成す角度は、等
角度に限らず任意に設定でき、また、直線の数は限定さ
れず、少なくとも照合指紋画像と登録指紋画像との照合
を行うために最低限必要な本数あればよい。The memory 20 detects the center position of the registered fingerprint image and detects the brightness of a plurality of straight lines passing through the detected center position with respect to the fingerprint image (hereinafter referred to as a registered fingerprint image) which is the target of the collation fingerprint image. A straight line B 1 -b 1 of a registered fingerprint image obtained by previously performing each process of signal detection and conversion of the detected light / dark signal into a time / frequency domain signal
The power component values of the time / frequency domain signal of ~ B n -b n are stored. Straight lines B 1 -b 1 to B n of this registered fingerprint image
-B n, like the linear A 1 -a 1 ~A n -a n collation fingerprint image, based on the horizontal straight line B 1 -b 1, which at equal angular range of 180 degrees clockwise These are N straight lines obtained by rotation. The angle formed by each of the adjacent straight lines is not limited to the equal angle, and can be set arbitrarily.The number of straight lines is not limited, and at least the minimum necessary for matching the matching fingerprint image with the registered fingerprint image is required. The number is sufficient.
【００２１】照合部１８では、このメモリ２０に記憶さ
れた登録指紋画像の直線Ｂ1−ｂ1〜Ｂn−ｂnに関する時
間・周波数領域信号のパワー成分値と、ウェーブレット
変換部１６で算出された照合指紋画像の直線Ａ1−ａ1〜
Ａn−ａnに関する時間・周波数領域信号のパワー成分値
とに基づいて算出した各々の直線の一致度に基づいて、
以下に示すように登録指紋画像と照合指紋画像とが一致
するか否かを判断する。なお、メモリ２０には、この判
断を行う制御ルーチンや所定閾値等のデータも記憶され
ている。The collating unit 18 calculates the power component values of the time / frequency domain signals regarding the straight lines B 1 -b 1 to B n -b n of the registered fingerprint image stored in the memory 20, and calculates the power component values by the wavelet transform unit 16. Straight line A 1 -a 1 to
Based on the respective straight lines matching degree calculated based on the power component values of the time-frequency domain signal regarding A n -a n,
As described below, it is determined whether the registered fingerprint image matches the collation fingerprint image. The memory 20 also stores a control routine for making this determination and data such as a predetermined threshold.
【００２２】次に、照合部１８で実行される制御ルーチ
ンについて図５に示すフローチャートを参照して説明す
る。Next, a control routine executed by the collating unit 18 will be described with reference to a flowchart shown in FIG.
【００２３】まず、照合部１８では、予めメモリ２０に
記憶されている登録指紋画像の直線に関する時間・周波
数領域信号の各々のパワー成分値を読み込み、ウェーブ
レット変換部１６で算出した照合指紋画像の直線Ａ1−
ａ1に関する時間・周波数領域信号のパワー成分値との
相互相関値を各々算出する（ステップ１００）。First, the collating unit 18 reads each power component value of the time / frequency domain signal relating to the straight line of the registered fingerprint image stored in the memory 20 in advance, and reads the straight line of the collated fingerprint image calculated by the wavelet transform unit 16. A 1 −
A cross-correlation value with respect to the power component value of the time / frequency domain signal relating to a 1 is calculated (step 100).
【００２４】すなわち、照合指紋画像の直線Ａ1−ａ1に
関する時間・周波数領域信号のパワー成分値と登録指紋
画像の直線Ｂ1−ｂ1に関するパワー成分値との相互相関
値を求め、次に照合指紋画像の直線Ａ1−ａ1に関する時
間・周波数領域信号のパワー成分値と登録指紋画像の直
線Ｂ2−ｂ2に関する時間・周波数領域信号のパワー成分
値との相互相関値を求める。以下同様にして、残りの直
線のそれぞれ相互相関値を求める。That is, the cross-correlation value between the power component value of the time / frequency domain signal on the straight line A 1 -a 1 of the collation fingerprint image and the power component value on the straight line B 1 -b 1 of the registered fingerprint image is obtained. A cross-correlation value between the power component value of the time / frequency domain signal on the straight line A 1 -a 1 of the collation fingerprint image and the power component value of the time / frequency domain signal on the straight line B 2 -b 2 of the registered fingerprint image is determined. Similarly, the cross-correlation values of the remaining straight lines are obtained in the same manner.
【００２５】この相互相関値の算出は、例えば、照合指
紋画像及び登録指紋画像のそれぞれの直線上の対応する
画素同士の積和値（内積）を求めることにより行う。ま
た、相互相関値を求める時の時間（画像では位置）の差
の範囲は、例えば、中心位置Ｘを原点とし、この原点を
中心として直線に沿った方向に±２０画素の範囲内の画
素とする。例えば、照合指紋画像の直線Ａ1−ａ1のパワ
ー成分値と直線Ｂ1−ｂ1を１画素ずらしたときのパワー
成分値との内積を求めることにより相互相関値を求め
る。同様に±２０画素の範囲内で直線Ｂ1−ｂ1のパワー
成分値を１画素ずつずらしたときの直線Ａ1−ａ1との相
互相関値を求める。The calculation of the cross-correlation value is performed, for example, by calculating the sum-of-product value (inner product) of the corresponding pixels on the straight lines of the verification fingerprint image and the registered fingerprint image. Further, the range of the difference in the time (position in the image) when the cross-correlation value is obtained is, for example, the center position X as the origin, and the pixels within the range of ± 20 pixels in the direction along the straight line with the origin as the center. I do. For example, the cross-correlation value is obtained by calculating the inner product of the power component value of the straight line A 1 -a 1 of the collation fingerprint image and the power component value obtained by shifting the straight line B 1 -b 1 by one pixel. Similarly, a cross-correlation value with the straight line A 1 -a 1 when the power component value of the straight line B 1 -b 1 is shifted by one pixel within the range of ± 20 pixels is obtained.
【００２６】図６（Ａ）に照合指紋画像の直線の明暗信
号と登録指紋画像の直線の明暗信号とが一致した場合の
例を、図６（Ｂ）に照合指紋画像の直線の明暗信号と登
録指紋画像の直線の明暗信号とが不一致の場合の例を示
す。図６（Ｂ）に示すように、照合指紋画像の直線の明
暗信号と登録指紋画像の直線の明暗信号とが不一致の場
合には、照合指紋画像の谷線部（又は隆線部）と登録指
紋画像の隆線部（又は谷線部）との信号が互いに打ち消
しあい、相互相関値はほぼ同じような値となる。FIG. 6A shows an example of the case where the straight-line light / dark signal of the collation fingerprint image coincides with the straight-line light / dark signal of the registered fingerprint image. FIG. An example of a case where the light and dark signals of the straight line of the registered fingerprint image do not match. As shown in FIG. 6 (B), when the light and dark signal of the straight line of the collation fingerprint image and the light and dark signal of the straight line of the registered fingerprint image do not match, the valley (or ridge) of the collation fingerprint image is registered. The signals with the ridges (or valleys) of the fingerprint image cancel each other, and the cross-correlation values are almost the same.
【００２７】次に、算出した各々の直線の相互相関値か
ら、各々の直線の一致度を算出する（ステップ１０
２）。この一致度Ｓは、ステップ１００で直線毎に求め
た各々の画素の相互相関値の最大値をＣｍａｘ、最小値
をＣｍｉｎとして（図６（Ａ）参照）、以下の（１）式
により求める。Next, the degree of coincidence of each straight line is calculated from the calculated cross-correlation value of each straight line (step 10).
2). The degree of coincidence S is determined by the following equation (1) with the maximum value of the cross-correlation value of each pixel determined for each straight line in step 100 as Cmax and the minimum value as Cmin (see FIG. 6A).
【００２８】 Ｓ＝（Ｃｍａｘ−Ｃｍｉｎ）／Ｃｍｉｎ ・・・（１） 次に、ステップ１０２で求めたそれぞれの直線の一致度
Ｓのうち、閾値Ｔ以上となる一致度Ｓが存在するか否か
を判断する（ステップ１０４）。そして、閾値Ｔ以上と
なる一致度Ｓが１つも存在しない場合には、照合指紋画
像と登録指紋画像とが不一致であると判断し、その旨を
図示しない表示装置等に出力し（ステップ１１６）、本
制御ルーチンを終了する。なお、上記と同様の処理を複
数回繰り返しても閾値Ｔ以上となる一致度Ｓが１つも存
在しない場合に本制御ルーチンを終了させるようにして
もよい。S = (Cmax−Cmin) / Cmin (1) Next, among the degrees of coincidence S of the respective straight lines obtained in step 102, whether or not there is a degree of coincidence S equal to or larger than the threshold value T Is determined (step 104). If there is no matching degree S equal to or greater than the threshold value T, it is determined that the matching fingerprint image and the registered fingerprint image do not match, and the fact is output to a display device or the like (not shown) (step 116). Then, the control routine ends. This control routine may be terminated when there is no matching degree S equal to or greater than the threshold value T even if the same processing as described above is repeated a plurality of times.
【００２９】一方、閾値Ｔ以上となる一致度Ｓが存在す
る場合には、照合指紋画像と登録指紋画像との回転ずれ
角度を算出する（ステップ１０６）。この回転ずれ角度
の算出は、算出した一致度Ｓが最大である直線の傾き角
度を求めることにより行う。すなわち、例えば直線の数
Ｎが１８の場合、照合指紋画像の直線Ａ1−ａ1と登録指
紋画像の直線Ｂ1−ｂ1との一致度Ｓが最大であった場合
には、回転ずれ角度は０度であり、照合指紋画像の直線
Ａ1−ａ1と登録指紋画像の直線Ｂ2−ｂ2との一致度Ｓが
最大であった場合には、回転ずれ角度は１０度である。On the other hand, if there is a matching degree S equal to or larger than the threshold value T, the rotational deviation angle between the verification fingerprint image and the registered fingerprint image is calculated (step 106). The calculation of the rotational deviation angle is performed by obtaining the inclination angle of the straight line having the maximum calculated coincidence S. That is, for example, when the number N of the straight lines is 18, and when the degree of coincidence S between the straight line A 1 -a 1 of the collation fingerprint image and the straight line B 1 -b 1 of the registered fingerprint image is the maximum, the rotational deviation angle Is 0 degrees, and when the degree of coincidence S between the straight line A 1 -a 1 of the collation fingerprint image and the straight line B 2 -b 2 of the registered fingerprint image is the maximum, the rotational shift angle is 10 degrees.
【００３０】次に、照合指紋画像の直線に関する時間・
周波数領域信号のパワー成分値とステップ１０６で求め
た回転ずれ角度分ずらした（補正した）登録指紋画像の
直線に関する時間周波数領域信号のパワー成分値との相
互相関値をそれぞれ求める（ステップ１０８）。すなわ
ち、回転ずれ角度が１０度であった場合には、照合指紋
画像の直線Ａ2−ａ2に関する時間・周波数領域信号のパ
ワー成分値と登録指紋画像の直線Ｂ3−ｂ3に関する時間
・周波数領域信号のパワー成分値との相互相関値を算出
し、照合指紋画像の直線Ａ3−ａ3に関する時間・周波数
領域信号のパワー成分値と登録指紋画像の直線Ｂ4−ｂ4
に関する時間・周波数領域信号のパワー成分値との相互
相関値を算出する。以下同様にしてＮ−１個分の相互相
関値を算出する。Next, the time for the straight line of the collation fingerprint image
A cross-correlation value between the power component value of the frequency domain signal and the power component value of the time frequency domain signal with respect to the straight line of the registered fingerprint image shifted (corrected) by the rotational shift angle obtained in step 106 is obtained (step 108). That is, when the rotational displacement angle is 10 degrees, the linear B 3 -b 3 with respect to time and frequency of the power component value and the registered fingerprint image of the time-frequency domain signal with respect to the linear A 2 -a 2 of the collation fingerprint image calculating a cross-correlation value between the power component value of the region signal, the collation fingerprint image linear a 3 linear power component value and the registered fingerprint image of the time-frequency domain signals relating -a 3 B 4 -b 4 of
A cross-correlation value with the power component value of the time / frequency domain signal is calculated. Hereinafter, the cross-correlation values for N−1 are calculated in the same manner.
【００３１】次に、算出した相互相関値から（１）式に
よりそれぞれの直線の一致度Ｓを算出する（ステップ１
１０）。そして、閾値Ｔ以上の一致度Ｓが所定個Ｙ以上
存在するか否かを判断する（ステップ１１２）。閾値Ｔ
以上の一致度Ｓが所定個Ｙ以上存在しない場合には、照
合指紋画像と登録指紋画像とが不一致であると判断し、
その旨を図示しない表示装置等に出力し（ステップ１１
６）、本制御ルーチンを終了する。なお、閾値Ｔ以上の
一致度Ｓが所定個Ｙ以上存在しない場合には、照合指紋
画像の直線Ａ1−ａ1との一致度Ｓが例えば２番目に大き
い登録指紋画像の直線の傾き角度を求めて回転ずれ角度
として上記の処理を再び行うようにしてもよい。Next, the degree of coincidence S of each straight line is calculated from the calculated cross-correlation value by the equation (1) (step 1).
10). Then, it is determined whether or not the predetermined number Y of the coincidence S equal to or greater than the threshold T exists (step 112). Threshold T
If the degree of coincidence S does not exceed the predetermined number Y, it is determined that the collation fingerprint image and the registered fingerprint image do not match,
This is output to a display device or the like (not shown) (step 11).
6), the control routine ends. If there is no matching degree S equal to or greater than the threshold value T equal to or more than the predetermined number Y, the matching degree S with the straight line A 1 -a 1 of the collation fingerprint image is, for example, the inclination angle of the straight line of the registered fingerprint image which is the second largest. The above-described processing may be performed again as the rotational deviation angle obtained.
【００３２】一方、閾値Ｔ以上の一致度Ｓが所定個Ｙ以
上存在する場合には、照合指紋画像と登録指紋画像とが
一致したと判断し、その旨を図示しない表示装置等に出
力し（ステップ１１４）、本制御ルーチンを終了する。On the other hand, if the number of coincidences S equal to or greater than the threshold value T is equal to or greater than the predetermined number Y, it is determined that the collation fingerprint image matches the registered fingerprint image, and the fact is output to a display device or the like (not shown). Step 114), this control routine ends.
【００３３】このように、指紋画像の中心位置を通る直
線上の画像（１次元の明暗信号）を時間・周波数領域へ
変換して処理することにより、照合指紋画像の位置ずれ
や回転ずれに影響されることなく、正確に照合すること
ができる。また、時間・周波数領域へ変換するための周
波数を隆線の平均ピッチに相当する周波数の２倍以上と
することにより、隆線部を強調して検出することがで
き、より正確に照合することができる。さらに、２次元
の画像を所定数の１次元の明暗信号に分割して処理する
ので、従来のように画素単位で各演算処理をする場合と
比較して処理時間を大幅に短縮することができる。As described above, by converting an image on a straight line (one-dimensional light / dark signal) passing through the center position of the fingerprint image into the time / frequency domain and processing the same, the position and rotation of the verification fingerprint image are affected. It is possible to perform accurate matching without being performed. Also, by setting the frequency for conversion to the time / frequency domain to be twice or more the frequency corresponding to the average pitch of the ridges, the ridges can be emphasized and detected, and more accurate matching can be performed. Can be. Further, since the two-dimensional image is divided into a predetermined number of one-dimensional light / dark signals and processed, the processing time can be greatly reduced as compared with the conventional case where each arithmetic processing is performed in pixel units. .
【００３４】なお、本実施の形態では、明暗信号にウェ
ーブレット変換を施して得られた時間・周波数領域信号
のパワー成分値に基づいて相互相関値を求めたが、これ
に限らず、図４（Ｃ）に示すような、明暗信号にウェー
ブレット変換を施して得られた時間・周波数領域信号の
位相成分に基づいて相互相関値を求めるようにしてもよ
い。また、パワー成分及び位相成分の両方に基づいて相
互相関値を求めるようにしてもよい。In the present embodiment, the cross-correlation value is obtained based on the power component value of the time / frequency domain signal obtained by performing the wavelet transform on the light / dark signal. However, the present invention is not limited to this. The cross-correlation value may be obtained based on the phase component of the time / frequency domain signal obtained by performing the wavelet transform on the light / dark signal as shown in C). Further, the cross-correlation value may be obtained based on both the power component and the phase component.
【００３５】[0035]
【発明の効果】以上説明したように、請求項１記載の発
明によれば、照合指紋画像を中心位置を通る予め定めた
複数の直線上の画素の明暗信号を検出し、これを時間・
周波数領域信号へ変換して処理するので、処理時間を大
幅に短縮することができると共に、照合指紋画像の位置
ずれに影響されることなく、正確に照合することができ
る、という効果を有する。As described above, according to the first aspect of the present invention, the light and dark signals of pixels on a plurality of predetermined straight lines passing through the center position of the collation fingerprint image are detected, and the signals are detected in time and time.
Since the signal is converted into a frequency domain signal and processed, the processing time can be greatly reduced, and there is an effect that accurate collation can be performed without being affected by the displacement of the collation fingerprint image.
【００３６】請求項２記載の発明によれば、照合指紋画
像と登録指紋画像とに回転ずれが有る場合でも正確に照
合することができる、という効果を有する。According to the second aspect of the present invention, there is an effect that accurate collation can be performed even when there is a rotational shift between the collation fingerprint image and the registered fingerprint image.
【図１】指紋照合装置の概略構成図である。FIG. 1 is a schematic configuration diagram of a fingerprint matching device.
【図２】（Ａ）〜（Ｄ）は指紋の紋様の一例を示す図で
ある。FIGS. 2A to 2D are diagrams illustrating an example of a fingerprint pattern.
【図３】指紋をの中心位置を通る直線の一例を示す図で
ある。FIG. 3 is a diagram illustrating an example of a straight line passing through a center position of a fingerprint.
【図４】（Ａ）は明暗信号を示す図、（Ｂ）はウェーブ
レット変換後の明暗信号のパワーを示す図、（Ｃ）ウェ
ーブレット変換後の明暗信号の位相を示す図である。4A is a diagram showing a light-dark signal, FIG. 4B is a diagram showing the power of the light-dark signal after wavelet transform, and FIG. 4C is a diagram showing the phase of the light-dark signal after wavelet transform.
【図５】照合部で実行される制御ルーチンの流れを示す
フローチャートである。FIG. 5 is a flowchart illustrating a flow of a control routine executed by a matching unit;
【図６】（Ａ）は、照合指紋及び登録指紋の明暗信号が
一致した場合の相互相関値を示す図、（Ｂ）は、照合指
紋及び登録指紋の明暗信号が不一致だった場合の相互相
関値を示す図である。FIG. 6A is a diagram showing a cross-correlation value when the light and dark signals of the collation fingerprint and the registered fingerprint match, and FIG. 6B is a cross-correlation when the light and dark signals of the collation fingerprint and the registered fingerprint do not match. It is a figure showing a value.
１０ 指紋照合装置 １２ 中心位置検出部（中心位置検出手段） １４ 明暗信号検出部（明暗信号検出手段） １６ ウェーブレット変換部（ウェーブレット変換手
段） １８ 照合部（一致度算出手段、照合手段） ２０ メモリDESCRIPTION OF SYMBOLS 10 Fingerprint collation device 12 Center position detection part (central position detection means) 14 Light / dark signal detection part (light / dark signal detection means) 16 Wavelet transformation part (wavelet transformation means) 18 Collation part (coincidence calculation means, collation means) 20 Memory
───────────────────────────────────────────────────── フロントページの続き Ｆターム(参考） 5B043 AA09 BA02 EA07 EA10 GA04 5L096 AA06 BA15 CA12 FA26 FA34 FA62 JA03 JA11 ────────────────────────────────────────────────── ─── Continued on the front page F term (reference) 5B043 AA09 BA02 EA07 EA10 GA04 5L096 AA06 BA15 CA12 FA26 FA34 FA62 JA03 JA11
Claims (4)
る中心位置検出手段と、 検出した中心位置を通る予め定めた複数の直線上の画素
の明暗信号を各々検出する明暗信号検出手段と、 検出した複数の明暗信号を予め定めた指紋の隆線の平均
ピッチに相当する周波数よりも高い周波数で時間・周波
数領域信号へ変換する変換手段と、 前記複数の直線の時間・周波数領域信号の所定成分と、
予め定めた登録指紋画像について該画像の紋様の中心位
置を通り、前記予め定めた複数の直線に相当する直線上
の画素列に関する時間・周波数領域信号の所定成分とに
基づいて各直線の相互相関値を算出し、該算出した相互
相関値に基づいて各直線の一致度を各々算出する一致度
算出手段と、 各々算出された一致度に基づいて前記照合指紋画像と前
記登録指紋画像との照合を行う照合手段と、 を有する指紋照合装置。1. A center position detecting means for detecting a center position of a pattern of a collation fingerprint image; a light / dark signal detecting means for detecting light / dark signals of pixels on a plurality of predetermined straight lines passing through the detected center position; Converting means for converting the plurality of detected light / dark signals into a time / frequency domain signal at a frequency higher than a frequency corresponding to a predetermined average pitch of fingerprint ridges; Ingredients and
The cross-correlation of each straight line based on a predetermined component of a time / frequency domain signal related to a pixel array on a straight line corresponding to the predetermined plurality of straight lines, passing through the center position of the pattern of the predetermined registered fingerprint image, Value calculating means for calculating the degree of coincidence of each straight line based on the calculated cross-correlation value; and collating the collated fingerprint image with the registered fingerprint image based on the computed degree of coincidence. And a fingerprint collation device comprising:
像の中心位置を通る複数の直線の中の予め定めた直線に
関する時間・周波数領域信号の所定成分と、前記登録指
紋画像の中心位置を通る前記予め定めた複数の直線に関
する時間・周波数領域信号の所定成分との相互相関値を
各々算出し、該算出した相互相関値が最も高い直線を基
準として他の各直線の一致度を各々算出することを特徴
とする請求項１に記載の指紋照合装置。2. The method according to claim 1, wherein the coincidence calculating unit calculates a predetermined component of a time / frequency domain signal relating to a predetermined straight line among a plurality of straight lines passing through a center position of the collation fingerprint image and a center position of the registered fingerprint image. A cross-correlation value of each of the plurality of predetermined straight lines passing through with a predetermined component of the time / frequency domain signal is calculated, and the degree of coincidence of each of the other straight lines is calculated based on the straight line having the highest calculated cross-correlation value. The fingerprint collation device according to claim 1, wherein the fingerprint collation is performed.
分の少なくとも一方であることを特徴とする請求項１又
は請求項２に記載の指紋照合装置。3. The fingerprint matching device according to claim 1, wherein the predetermined component is at least one of a power component and a phase component.
ことを特徴とする請求項１乃至請求項３の何れか１項に
記載の指紋照合装置。4. The fingerprint matching device according to claim 1, wherein the transform is a wavelet transform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP11162290A JP2000348178A (en) | 1999-06-09 | 1999-06-09 | Fingerprint collation device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP11162290A JP2000348178A (en) | 1999-06-09 | 1999-06-09 | Fingerprint collation device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JP2000348178A true JP2000348178A (en) | 2000-12-15 |
Family
ID=15751685
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP11162290A Pending JP2000348178A (en) | 1999-06-09 | 1999-06-09 | Fingerprint collation device |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP2000348178A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005312748A (en) * | 2004-04-30 | 2005-11-10 | Sony Corp | Identification device and method, and authentication system |
| WO2013172413A1 (en) * | 2012-05-18 | 2013-11-21 | Ｎｅｃソフト株式会社 | Fingerprint ridge image synthesis system, fingerprint ridge image synthesis method, and program thereof |
| JP2015026283A (en) * | 2013-07-26 | 2015-02-05 | 富士通株式会社 | Image processing apparatus, image processing method, and program |
-
1999
- 1999-06-09 JP JP11162290A patent/JP2000348178A/en active Pending
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005312748A (en) * | 2004-04-30 | 2005-11-10 | Sony Corp | Identification device and method, and authentication system |
| JP4636306B2 (en) * | 2004-04-30 | 2011-02-23 | ソニー株式会社 | Identification device, identification method, and authentication system |
| WO2013172413A1 (en) * | 2012-05-18 | 2013-11-21 | Ｎｅｃソフト株式会社 | Fingerprint ridge image synthesis system, fingerprint ridge image synthesis method, and program thereof |
| CN104303209A (en) * | 2012-05-18 | 2015-01-21 | 日本电气方案创新株式会社 | Fingerprint ridge image synthesis system, fingerprint ridge image synthesis method, and program thereof |
| KR20150020327A (en) * | 2012-05-18 | 2015-02-25 | 엔이씨 솔루션 이노베이터 가부시키가이샤 | Fingerprint ridge image synthesis system, fingerprint ridge image synthesis method, and program thereof |
| US20150161432A1 (en) * | 2012-05-18 | 2015-06-11 | Nec Solution Innovators, Ltd. | Fingerprint ridge image synthesis system, fingerprint ridge image synthesis method, and program thereof |
| JPWO2013172413A1 (en) * | 2012-05-18 | 2016-01-12 | Ｎｅｃソリューションイノベータ株式会社 | Fingerprint ridge image composition system, fingerprint ridge image composition method, and program thereof |
| US9323974B2 (en) | 2012-05-18 | 2016-04-26 | Nec Solution Innovators, Ltd. | Fingerprint ridge image synthesis system, fingerprint ridge image synthesis method, and program thereof |
| KR101627062B1 (en) | 2012-05-18 | 2016-06-03 | 엔이씨 솔루션 이노베이터 가부시키가이샤 | Fingerprint ridge image synthesis system, fingerprint ridge image synthesis method, and program thereof |
| CN104303209B (en) * | 2012-05-18 | 2017-05-17 | 日本电气方案创新株式会社 | Fingerprint ridge image synthesis system and fingerprint ridge image synthesis method |
| JP2015026283A (en) * | 2013-07-26 | 2015-02-05 | 富士通株式会社 | Image processing apparatus, image processing method, and program |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6144754A (en) | Method and apparatus for identifying individuals | |
| US7496214B2 (en) | Method of palm print identification | |
| US7970186B2 (en) | System, method and computer program product for fingerprint verification | |
| KR101314945B1 (en) | Fake finger determination device | |
| US7512257B2 (en) | Coding system and method of a fingerprint image | |
| KR100374708B1 (en) | Non-contact type human iris recognition method by correction of rotated iris image | |
| US6961453B2 (en) | Method for extracting fingerprint feature data using ridge orientation model | |
| US20060023921A1 (en) | Authentication apparatus, verification method and verification apparatus | |
| El-Sisi | Design and implementation biometric access control system using fingerprint for restricted area based on gabor filter. | |
| Sagayam et al. | Authentication of biometric system using fingerprint recognition with euclidean distance and neural network classifier | |
| Tukur | Fingerprint recognition and matching using Matlab | |
| WO2008143271A1 (en) | Registering device, checking device, program, and data structure | |
| US5684894A (en) | Scale specific and robust line/edge encoding of images | |
| JP2000348178A (en) | Fingerprint collation device | |
| Biradar et al. | Fingerprint ridge orientation extraction: a review of state of the art techniques | |
| JP2903788B2 (en) | Fingerprint image input device | |
| JP2000003436A (en) | Device and method for recognizing isar picture | |
| JPH0755429A (en) | Position recognition system | |
| JPH08287259A (en) | Fingerprint identifying method | |
| Prakram et al. | EVALUATION OF IMPROVED FUZZY INFERENCE SYSTEM TO PRESERVE IMAGE EDGE FOR IMAGE ANALYSIS. | |
| JP2007179267A (en) | Pattern matching device | |
| JPH01271884A (en) | Detecting system for center of fingerprint | |
| Wieclaw | Fingerprint orientation field enhancement | |
| Kovac et al. | Multimodal biometric system based on fingerprint and finger vein pattern | |
| JPH09114990A (en) | Image recognizing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| R250 | Receipt of annual fees |
Free format text: JAPANESE INTERMEDIATE CODE: R250 |
|
| LAPS | Cancellation because of no payment of annual fees |