JPH01271884A - Detecting system for center of fingerprint - Google Patents
Detecting system for center of fingerprintInfo
- Publication number
- JPH01271884A JPH01271884A JP63099335A JP9933588A JPH01271884A JP H01271884 A JPH01271884 A JP H01271884A JP 63099335 A JP63099335 A JP 63099335A JP 9933588 A JP9933588 A JP 9933588A JP H01271884 A JPH01271884 A JP H01271884A
- Authority
- JP
- Japan
- Prior art keywords
- curvature
- curve
- center
- fingerprint
- minimum
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 claims description 15
- 239000000284 extract Substances 0.000 abstract description 4
- 230000004044 response Effects 0.000 abstract 2
- 230000000630 rising effect Effects 0.000 abstract 1
- 238000000605 extraction Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000000034 method Methods 0.000 description 4
- 238000012937 correction Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
Description
【発明の詳細な説明】
〔産業上の利用分野〕
本発明は指紋の照合に用いられる照合装置等において、
指紋画像上の各点を座標で表わす際必要な指紋中心を決
定するための指紋中心検出方式に関する。[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a verification device used for fingerprint verification, etc.
The present invention relates to a fingerprint center detection method for determining the fingerprint center required when representing each point on a fingerprint image with coordinates.
一般に指紋の同定を行う際には、指紋紋様上における特
徴点（例えば端点や分岐点）の位置を求め、これら特徴
点の位置の一致、不一致によって指紋の同定を行ってい
る。近年、この指紋の同定を画像処理技術を利用して、
コンビーータで行なうことが実現されている。Generally, when identifying a fingerprint, the positions of minutiae (for example, end points or branch points) on the fingerprint pattern are determined, and the fingerprint is identified based on whether or not the positions of these minutiae points match or differ. In recent years, image processing technology has been used to identify fingerprints.
What can be done with a conveter has been realized.
従来、この特徴点の位置決定の際、利用される指紋中心
を求める場合は、指紋隆線を指紋上部に位置するものか
ら順次トレースし、それぞれの隆線の頂上点を検出し、
最後に頂上点が検出されだ隆線が指紋紋様を構成する隆
線の最も内側の隆線とみなされ、この隆線の検出された
頂上点を指紋中心としている（例えば特公昭５８−５５
５４９号公報）。Conventionally, when determining the position of minutiae, to find the center of the fingerprint used, the fingerprint ridges are traced sequentially starting from the one located at the top of the fingerprint, and the top point of each ridge is detected.
The ridge whose peak point is finally detected is regarded as the innermost ridge of the ridges that make up the fingerprint pattern, and the detected peak point of this ridge is the center of the fingerprint (for example,
549 Publication).
ところで、上述の指紋中心検出方式の場合、第５図（、
）に示す蹄状紋様及び第５図（ｂ）に示す渦状紋様の指
紋には対応できるが、第５図（ｃ）に示す弓状紋様の指
紋の場合、最も内側の隆線が存在しないため、原理的に
指紋中心の検出ができないという問題点がある。By the way, in the case of the above-mentioned fingerprint center detection method, Fig. 5 (,
) and the spiral pattern shown in Figure 5(b), but in the case of the arcuate pattern shown in Figure 5(c), the innermost ridge does not exist. However, there is a problem in that, in principle, it is not possible to detect the center of the fingerprint.
さらに、従来の指紋中心検出方式では、第５図（ｄ）に
示すように、その中心部が不鮮明な指紋画像及び第５図
（ｅ）に示すようにその中心部上部が不鮮明な指紋画像
の場合、隆線のトレースができず。Furthermore, in the conventional fingerprint center detection method, as shown in FIG. 5(d), the center of the fingerprint image is unclear, and as shown in FIG. 5(e), the upper part of the center is unclear. In this case, the ridge cannot be traced.
その結果、指紋中心の検出できないという問題点がある
。As a result, there is a problem that the center of the fingerprint cannot be detected.
従って、実際の指紋画像では低品質の指紋画像が多いだ
め、多くの指紋画像で、指紋中心の検出が困難となる。Therefore, since there are many low-quality fingerprint images in actual fingerprint images, it is difficult to detect the center of the fingerprint in many fingerprint images.
まだ、指紋中心が検出された場合でもオ被レークによっ
て、検出指紋中心の確認。Even if the fingerprint center is still detected, the detected fingerprint center can be confirmed by checking the detection area.
修正が必要となり、オペレータによる確認修正のだめの
時間が極めて多くなってしまう。Corrections are required, and the amount of time it takes for the operator to check and make corrections becomes extremely large.
本発明の目的は高精度に指紋中心を検出することができ
る指紋中心検出方式を提供することにある。An object of the present invention is to provide a fingerprint center detection method that can detect the center of a fingerprint with high accuracy.
本発明の指紋中心検出方式は９画像各絵素における指紋
隆線の方向および曲率を２次元記憶する方向曲率記憶部
と、上に凸な形状を持つ隆線の頂上点近傍における曲率
を、垂直方向の座標（Ｙ座標）毎に一時的に記憶する上
向き曲率一時記憶部と。The fingerprint center detection method of the present invention includes a direction curvature storage unit that two-dimensionally stores the direction and curvature of a fingerprint ridge in each pixel of nine images, and a curvature near the top of a ridge that has an upwardly convex shape. an upward curvature temporary storage unit that temporarily stores each direction coordinate (Y coordinate);
種々の典型的な指紋についての垂直方向の曲率変化向−
とこの曲率変化曲線に対応する中心の想定位置が予め記
憶された曲率曲線パターン辞書と、入力指紋の各Ｙ座標
値の上向き曲率を各曲率曲線ｉｅターンに近似させた場
合の曲線誤差の最小値とこの曲線誤差の最小値に対応す
る想定中心Ｙ座標値を一時的に記憶する曲線誤差一時記
憶部とを備えるとともに、各Ｙ座標に対応して前記方向
曲率記憶部から引き出され、Ｙ方向に直交する水平方向
の各絵素の曲率の中での最大の上向き曲率を前記上向き
曲率一時記憶部に登録する上向き曲率抽出部と、前記上
向き曲率一時記憶部に登録されている上向き曲率列を。Vertical curvature change direction for various typical fingerprints
and a curvature curve pattern dictionary in which the assumed position of the center corresponding to this curvature change curve is stored in advance, and the minimum value of the curve error when the upward curvature of each Y coordinate value of the input fingerprint is approximated to each curvature curve ie turn. and a curve error temporary storage section that temporarily stores the assumed center Y coordinate value corresponding to the minimum value of the curve error, and a curve error temporary storage section that temporarily stores the assumed center Y coordinate value corresponding to the minimum value of the curve error, and a curve error that is pulled out from the direction curvature storage section corresponding to each Y coordinate, and in the Y direction. an upward curvature extraction unit that registers a maximum upward curvature among the curvatures of each picture element in orthogonal horizontal directions in the upward curvature temporary storage unit; and an upward curvature string registered in the upward curvature temporary storage unit.
前記曲率曲線パターン辞書に登録されている各曲率曲線
パターンに近似させた際の曲線誤差最小値とこの曲線誤
差最小値に対応する想定中心Ｙ座標値を前記曲率誤差一
時記憶部に登録する曲率曲線Ｃ５）
誤差算出部と、前記曲率誤差一時記憶部に登録されてい
る各曲率曲線パターンの曲線誤差最小値の中での最小値
とこの最小値に対応する想定中心Ｙ座標を出力する中心
Ｙ座標検出部を有することを特徴としており、これによ
って弓状紋様の指紋及び指紋中心部が不鮮明な指紋であ
っても正確な中心のＹ座標を検出することができる。A curvature curve that registers a minimum curve error value when approximating each curvature curve pattern registered in the curvature curve pattern dictionary and an assumed center Y coordinate value corresponding to this minimum curve error value in the curvature error temporary storage unit. C5) An error calculation unit and a center Y coordinate that outputs the minimum value among the minimum curve error values of each curvature curve pattern registered in the curvature error temporary storage unit and the assumed center Y coordinate corresponding to this minimum value. The present invention is characterized in that it has a detection section, so that even if the fingerprint has an arcuate pattern or the center of the fingerprint is unclear, it is possible to accurately detect the Y coordinate of the center.
以下本発明について図面を参照して説明する。 The present invention will be explained below with reference to the drawings.
まず、第１図（ａ）〜（ｇ）を参照して本発明の原理に
ついて説明する。ここで第１図（、）には、蹄状紋を例
として、指紋画像を水平方向（Ｘ方向）にみた際、上側
に凸曲線の最大曲率が垂直方向に示されている。第１図
（ｂ）は第１図（ａ）を離散グラフに表したものである
。また、第１図（Ｃ）には弓状紋を例として、上向き曲
率が示されておシ、第１図（ｄ）はこの弓状紋の例を離
散グラフに表わしたものである。First, the principle of the present invention will be explained with reference to FIGS. 1(a) to 1(g). Here, in FIG. 1 (,), taking a hoof-like pattern as an example, when a fingerprint image is viewed in the horizontal direction (X direction), the maximum curvature of the convex curve on the upper side is shown in the vertical direction. FIG. 1(b) is a discrete graph representation of FIG. 1(a). Further, FIG. 1(C) shows an upward curvature of an arcuate pattern as an example, and FIG. 1(d) shows an example of this arcuate pattern in a discrete graph.
さらに第１図（ｅ） ｌ （ｆ）及び（ｇ）には上向き
曲率曲線の典型的なパターンの例が示されている。Furthermore, examples of typical patterns of upward curvature curves are shown in FIGS. 1(e), (f) and (g).
第１図（ａ）を参照すると、上向きの曲率は、指紋上部
から指紋中心部にかけて順次大きくなり、指紋中心部で
最大値となる。そして指紋中心部よシ下方では、上に凸
の曲線が抽出でき々くなる。上に凸の形状の指紋隆線の
内で、想定できる最大の曲率をＭｅと定義すると、すべ
ての指紋隆線について、−Ｌ向き曲″Ｊ′、は、指紋上
部から指紋中心部にかけて想定曲率最大値Ｍｅに近づい
ていく。上向き曲線を抽出できない中心部下方の領域に
対して想定曲率最大値Ｍｅよシ大きな値を設定すると、
上向き曲率は、指紋上部から指紋下部にかけて単調に増
加し、中心部直後に想定曲率最大値Ｍｅを超える。Referring to FIG. 1(a), the upward curvature gradually increases from the top of the fingerprint to the center of the fingerprint, and reaches its maximum value at the center of the fingerprint. Further, in the lower part of the fingerprint center, it becomes difficult to extract upwardly convex curves. If the maximum curvature that can be assumed among the fingerprint ridges with an upwardly convex shape is defined as Me, then for all the fingerprint ridges, the -L direction curve "J'" is the expected curvature from the top of the fingerprint to the center of the fingerprint. It approaches the maximum value Me.If a value larger than the assumed maximum curvature Me is set for the area below the center where an upward curve cannot be extracted,
The upward curvature increases monotonically from the upper part of the fingerprint to the lower part of the fingerprint, and exceeds the assumed maximum curvature value Me immediately after the center.
第１図（ｂ）に示すように中心部領域が鮮明で曲率抽出
が可能な場合は、想定曲率最大値Ｍｅを超す直前の点の
Ｙ座標（垂直方向の座標）をとれば正確な中心のＹ座標
と同一であることがわかる。If the center area is clear and curvature extraction is possible as shown in Figure 1(b), the exact center can be found by taking the Y coordinate (vertical coordinate) of the point just before the assumed maximum curvature value Me. It can be seen that it is the same as the Y coordinate.
次に第１図（ｃ）及び（ｄ）を参照すると、弓状紋の例
では２曲率近似曲線は山型になることがわかる。Next, referring to FIGS. 1(c) and 1(d), it can be seen that in the case of an arcuate pattern, the two-curvature approximate curve has a chevron shape.
中心部領域の曲率が抽出されている場合、上向き曲率の
最大値をとれば、弓状紋においても正確な中心のＹ座標
が検出される。When the curvature of the central area is extracted, if the maximum value of the upward curvature is taken, the accurate Y coordinate of the center can be detected even in the arcuate pattern.
第１図（ｅ）、（ｆ）及び（ｇ）を参照すると、与えら
れた指紋の上向き曲率が抽出されると、典型パターン群
の各々の曲線と比較して２曲線誤差が最小となるものの
想定中心位置を用いて指紋中心のＹ座標を検出すれば、
正確な中心を得る。Referring to FIGS. 1(e), (f), and (g), when the upward curvature of a given fingerprint is extracted, the two-curve error is minimized when compared with each curve of the typical pattern group. If the Y coordinate of the fingerprint center is detected using the assumed center position,
Get exact center.
この曲線近似を用いて中心検出を行なうと、第４図（ａ
）及び（ｃ）に示すように指紋中心部が不鮮明で曲率が
抽出できない場合でもあって、それぞれ第４図（ｂ）及
び（ｄ）の曲率曲線パターンと一致させることによって
正確な指紋中心Ｙ座標が検出できる。When the center is detected using this curve approximation, Figure 4 (a
) and (c), even if the center of the fingerprint is unclear and the curvature cannot be extracted, the exact Y coordinate of the center of the fingerprint can be determined by matching the curvature curve patterns in FIGS. 4(b) and (d), respectively. can be detected.
次に９本発明について第２図を参照して詳細に説明する
。Next, the present invention will be explained in detail with reference to FIG.
本発明による指紋中心検出方式では９画像各絵素におけ
る指紋隆線の方向及び曲率を２次元記憶する方向曲率記
憶部１２と、予め定められた方向（例えば上側）に凸の
形状を有する指紋隆線の頂上点近傍における曲率を予め
定められた方向の座標（垂直方向の座標、Ｙ座標）毎に
一時的に記憶する上向き曲率一時記憶部１３と、多数の
典型的な指紋についての垂直方向の曲率変化曲線とこの
曲率変化曲線に対応する中心の想定位置が予め記憶され
ている曲率曲線パターン辞書１４と、入力指紋の各Ｙ座
標値の上向き曲率を各曲率曲線パターンに近似させた際
の曲線誤差の最小値及びこの最小値に対応する想定中心
Ｙ座標値を一時的に記憶する曲線誤差一時記憶部１５と
が備えられており、さらに、各Ｙ座標に対応して水平方
向における各絵素の方向曲率を方向曲率記憶部１２から
引き出して、最大の上向き曲率を抽出し、この最大上向
き曲率を上向き曲率一時記憶部１３に登録する上向き曲
率抽出部１６と、入力画像の上向き曲率列を２曲率曲線
パターン辞書１４に登録されている各曲率曲線パターン
に近似させ２曲線誤差の最小値と、その最小値における
想定中心Ｙ座標値を算出し２曲線誤差最小値と想定中心
Ｙ座標とを曲率誤差一時記憶部１５に登録する曲率曲線
誤差算出部１７と、各曲率曲線パターンの曲線誤差最小
値の中での最小値を検出し、その最小値に対応する想定
中心Ｙ座標を出力する中心Ｙ座標検出部１８と、上向き
曲率抽出部１６．曲率曲線誤差算出部１７．及び中心Ｙ
座標検出部１８を制御する制御部１１が備えられている
。The fingerprint center detection method according to the present invention includes a direction curvature storage unit 12 that two-dimensionally stores the direction and curvature of a fingerprint ridge in each pixel of nine images, and a fingerprint ridge having a convex shape in a predetermined direction (for example, upward). An upward curvature temporary storage unit 13 temporarily stores the curvature near the top point of the line for each predetermined direction coordinate (vertical direction coordinate, Y coordinate), and an upward curvature temporary storage unit 13 that temporarily stores the curvature in the vicinity of the top point of the line for each predetermined direction coordinate (vertical direction coordinate, Y coordinate), and A curvature curve pattern dictionary 14 in which curvature change curves and assumed positions of centers corresponding to these curvature change curves are stored in advance, and curves obtained when the upward curvature of each Y coordinate value of the input fingerprint is approximated to each curvature curve pattern. A curve error temporary storage section 15 that temporarily stores the minimum value of the error and the assumed center Y coordinate value corresponding to this minimum value is provided, and furthermore, each pixel in the horizontal direction is stored in correspondence with each Y coordinate. an upward curvature extraction section 16 that extracts the directional curvature of the input image from the directional curvature storage section 12, extracts the maximum upward curvature, and registers this maximum upward curvature in the upward curvature temporary storage section 13; Approximate each curvature curve pattern registered in the curvature curve pattern dictionary 14 to calculate the minimum value of the two-curve error and the assumed center Y-coordinate value at that minimum value, and calculate the two-curve error minimum value and the assumed center Y-coordinate value as the curvature. A curvature curve error calculation unit 17 that is registered in the error temporary storage unit 15 and a center Y that detects the minimum value among the minimum curve error values of each curvature curve pattern and outputs the assumed center Y coordinate corresponding to the minimum value. Coordinate detection section 18 and upward curvature extraction section 16. Curvature curve error calculation unit 17. and center Y
A control section 11 that controls the coordinate detection section 18 is provided.
第３図（ａ）及び（ｂ）、に第１図（ａ）に示す指紋画
像の各絵素の方向曲率を表現する際の基準の一例を倒す
。FIGS. 3(a) and 3(b) show an example of the standard for expressing the directional curvature of each picture element of the fingerprint image shown in FIG. 1(a).
この例では第１図（、）に示す８方向のそれぞれについ
て第１図（ｂ）に示すように７種類の曲率が定義されて
いる。第３図（ａ）の方向と第３図（ｂ）の曲率を組み
合わせると種々の曲線が方向６２曲率Ｃを用いてＡｄ、
ｃと定義される。ここで、第３図（ｃ）及び（ｄ）に曲
線の定義の例を示す。なお、量子化された指紋画像の方
向曲率を抽出するとして２例えば特公昭５２−９７２５
８公報及び特公昭５５−１３８１７４公報記載のものが
知られている。In this example, seven types of curvatures are defined as shown in FIG. 1(b) for each of the eight directions shown in FIG. 1(,). By combining the direction in Figure 3(a) and the curvature in Figure 3(b), various curves can be obtained using the direction 62 curvature C, Ad,
It is defined as c. Here, examples of curve definitions are shown in FIGS. 3(c) and 3(d). In addition, when extracting the directional curvature of a quantized fingerprint image, 2, for example, Japanese Patent Publication No. 52-9725
8 and Japanese Patent Publication No. 55-138174 are known.
方向曲率記憶部１２には、入力指紋画像の各絵素ににお
ける方向曲率が予め登録されており、−方２曲率曲線パ
ターン辞書１４には、第１図（ｅ）。The directional curvature of each picture element of the input fingerprint image is registered in advance in the directional curvature storage unit 12, and the curvature of the direction shown in FIG.
（ｆ）及び（ｇ）に例示される典型的な指紋の曲率変化
曲線とその想定中心位置が予め登録されている。Typical fingerprint curvature change curves and their assumed center positions illustrated in (f) and (g) are registered in advance.
なお、典型的な曲率曲線パターンの採取の際には１種々
の紋様について例えば、第５図（ａ） 、 （ｂ）及び
（ｃ）に示す典型的な指紋を選択し、それぞれの指紋の
上向き曲率を抽出して、第１図（ｂ）及び（ｄ）に示す
ようにグラフ化する。次に、これを例えば最小自乗法で
曲線近似し、これを一つの曲率曲線パターンとする。次
にこの曲率曲線パターンに対応する指紋中心の実際のＹ
座標値Ｙｃをその想定中心位置とすればよい。このよう
にして採取された曲率曲線−ぐターンは、第１図（ｅ）
、 （ｆ）及び（ｇ）に示すように上向き曲率をＣと
すれば、関数ｆを用いて。When collecting typical curvature curve patterns, for example, select typical fingerprints shown in Figures 5 (a), (b), and (c) for each pattern, and The curvature is extracted and graphed as shown in FIGS. 1(b) and (d). Next, this is approximated to a curve using, for example, the least squares method, and this is made into one curvature curve pattern. Next, the actual Y at the center of the fingerprint corresponding to this curvature curve pattern
The coordinate value Yc may be used as the assumed center position. The curvature curve sampled in this way is shown in Figure 1(e).
, As shown in (f) and (g), if the upward curvature is C, then using the function f.
Ｃ＝ ｆ （Ｙ） と表現できる。C = f (Y) It can be expressed as
上向き曲率抽出部１６は、各Ｙ座標離散値Ｙｎ；（ｎ＝
１，２．・・・、Ｎ）について水平方向（Ｘ軸方向）の
各絵素（Ｘｍ 、 Ｙｎ）；（ｍ＝１．２．−、Ｍ）の
方向曲率を方向曲率記憶部１２力・ら取り出し、その中
で最大の上向き曲率を検出し、それをＣｎとして上向き
曲率一時記憶部１３に登録する。The upward curvature extraction unit 16 extracts each Y-coordinate discrete value Yn; (n=
1, 2. ..., N), the directional curvature of each picture element (Xm, Yn); (m=1.2.-, M) in the horizontal direction (X-axis direction) is retrieved from the directional curvature storage unit 12, and the The maximum upward curvature among them is detected and registered in the upward curvature temporary storage section 13 as Cn.
曲率関数誤差算出部１７は、上向き曲率Ｃｎ ： （ｎ
＝１ 、２、−Ｎ）を２曲率曲線パターン辞書工４に登
録されている曲率曲線パターンの一つと比較し９曲線誤
差を最小となる曲線パターンの想定中心から導かれる中
心Ｙ座標を検出する。The curvature function error calculation unit 17 calculates the upward curvature Cn: (n
= 1, 2, -N) with one of the curvature curve patterns registered in the curvature curve pattern dictionary engineer 4, and detect the center Y coordinate derived from the assumed center of the curve pattern that minimizes the curve error. .
ここで入力指紋の土向き曲率をＣｎ ： （ｎ＝１．２
．・・・Ｎ）とし、ある曲率曲線パターンを
Ｃ＝ｆ（Ｙ）
とすると、最小自乗法を用いて、誤差ｄは。Here, the curvature of the input fingerprint is Cn: (n=1.2
．． ...N), and a certain curvature curve pattern is C=f(Y). Using the least squares method, the error d is.
と表現できる。It can be expressed as
次にＹ方向に曲率曲線パターンをαたけ平行移動した場
合の誤差ｄ（α）は
と表現できる。Next, the error d(α) when the curvature curve pattern is translated by α in the Y direction can be expressed as follows.
このｄ（α）が、αに関する最／ＪＱ値ｍｉｎ ｄ（α
）をとるときのαをα′とすると入力指紋の中心Ｙ座標
の推定値は（Ｙｃ＋α）となる。このように算出された
各曲率曲線パターンの最小誤差と最小誤差に対応する中
心Ｙ座標の推定値が２曲線誤差一時記憶部１５に登録さ
れる。This d(α) is the maximum/JQ value min d(α
), the estimated value of the center Y coordinate of the input fingerprint is (Yc+α). The minimum error of each curvature curve pattern calculated in this way and the estimated value of the center Y coordinate corresponding to the minimum error are registered in the two-curve error temporary storage section 15.
中心Ｙ座標検出部１８は９曲線誤差一時記憶部１５に登
録されている曲率曲線パターン間での最小誤差を求め、
この最小誤差に対応する曲率曲線パターンに対応する中
心Ｙ座標推定値を中心Ｙ座標として出力する。The center Y coordinate detection unit 18 finds the minimum error between the curvature curve patterns registered in the nine curve error temporary storage unit 15,
The estimated value of the center Y coordinate corresponding to the curvature curve pattern corresponding to this minimum error is output as the center Y coordinate.
なお、この実施例では量子化画像の各絵素毎に方向曲率
が定義されていることを前提としだが。Note that this embodiment assumes that the directional curvature is defined for each picture element of the quantized image.
複数の絵素を含む区画領域毎に方向曲率を定義する方法
も中心検出処理高速化の観点から有効である。A method of defining the directional curvature for each divided area including a plurality of picture elements is also effective from the viewpoint of speeding up the center detection process.
以上説明したように２本発明によれば、弓状紋及び中心
部が不鮮明な指紋に対しても中心検出が可能となり、そ
の結果、オペレータの中心確認修正工数を不要とし、照
合精度も向上させることが可能になる。As explained above, according to the present invention, the center can be detected even for arcuate prints and fingerprints whose centers are unclear.As a result, the operator's man-hours for checking and correcting the center are not required, and matching accuracy is also improved. becomes possible.
第１図（ａ） 、 （ｂ） 、 （ｃ） ｌ （ｄ）
ｌ （ｅ） ＋ ｆ句、及び（ｇ）は２本発明の詳細な
説明するだめの図、第２図は２本発明の一実施例を示す
ブロック図、第３図（ａ） ｚ （ｂ）、（Ｃ） 。
及び（ｄ）は２本発明の一実施例で使用される方向曲率
の例を示す図、第４図（ａ） ｓ （ｂ） ｓ （ｃ）
＊及び（ｄ）は本発明の一実施例で使用される中心部
不鮮明指紋の曲率変化曲線の例を示す図、第５図（ａ）
ｊ （ｂ） ｅ （ｃ） 。
（ｄ） ｔ （ｅ）及び（ｆ）は指紋の種類を説明する
ための図である。
１に制御部、１２二方向曲率記憶部、１３：上向き曲率
一時記憶部、１４：曲率曲線パターン辞書、１５：曲線
誤差一時記憶部、１６：上向き曲率抽出部、１７：曲率
曲線誤差算出部、１８：中心部Ｙ座標検出部。
区
〜ｊ
区
第５
（ａ）
忙）
（ｂ）
（ｄ）
（ｆ）Figure 1 (a), (b), (c) l (d)
l (e) + f clause, and (g) are 2 diagrams for detailed explanation of the present invention, Figure 2 is a block diagram showing an embodiment of the present invention, Figure 3 (a) z (b ), (C). and (d) are two diagrams showing examples of directional curvature used in one embodiment of the present invention, Figure 4 (a) s (b) s (c)
* and (d) are diagrams showing examples of curvature change curves of centrally blurred fingerprints used in one embodiment of the present invention, FIG. 5(a)
j (b) e (c). (d) t (e) and (f) are diagrams for explaining types of fingerprints. 1 a control section, 12 a two-way curvature storage section, 13: an upward curvature temporary storage section, 14: a curvature curve pattern dictionary, 15: a curve error temporary storage section, 16: an upward curvature extraction section, 17: a curvature curve error calculation section, 18: Center Y coordinate detection unit. Ward~j Ward 5 (a) Busy) (b) (d) (f)
Claims (1)
紋様中心を決定する際に用いられ、前記指紋画像各絵素
における指紋隆線の方向及び曲率を２次元記憶する方向
曲率記憶部と、予め定められた方向に凸な形状を有する
隆線の頂上点近傍における曲率を所定方向で規定された
座標毎に一時的に記憶する曲率一時記憶部と、複数の指
紋について前記所定方向の曲率変化曲線及び該曲率変化
曲線に対応する中心の想定位置が予め記憶された曲率曲
線パターン辞書と、入力指紋の前記座標値の曲率を前記
各曲率曲線パターンに近似させた際における曲線誤差の
最小値及び該最小値に対応する想定中心座標値を一時的
に記憶する曲線誤差一時記憶部とを備えるとともに、前
記座標値に対応して前記方向曲率記憶部から引き出され
、前記所定方向に直交する方向における各絵素の曲率の
中での最大の曲率を前記曲率一時記憶部に登録する上向
き曲率抽出部と、前記上向き曲率一時記憶部に登録され
ている曲率列を前記曲率曲線パターン辞書に登録されて
いる各曲率曲線パターンに近似させた際の曲線誤差最小
値及び該曲線誤差最小値に対応する想定中心座標値を前
記曲率誤差一時記憶部に登録する曲率曲線誤差算出部と
、前記曲率誤差一時記憶部に登録されている各曲率曲線
パターンの曲線誤差最小値の中での最小値と該最小値に
対応する想定中心座標を出力する中心座標検出部とを有
することを特徴とする指紋中心検出方式。1. A direction curvature storage unit that is used to determine the center of a fingerprint pattern of a fingerprint image quantized into a two-dimensional array of pixels, and stores two-dimensionally the direction and curvature of a fingerprint ridge in each pixel of the fingerprint image. a curvature temporary storage unit that temporarily stores the curvature near the top point of a ridge having a convex shape in a predetermined direction for each coordinate specified in a predetermined direction; a curvature curve pattern dictionary in which curvature change curves and assumed positions of centers corresponding to the curvature change curves are stored in advance; and a minimum curve error when approximating the curvature of the coordinate values of the input fingerprint to each of the curvature curve patterns. a curve error temporary storage section that temporarily stores a value and an assumed center coordinate value corresponding to the minimum value; an upward curvature extracting unit that registers a maximum curvature among the curvatures of each picture element in the direction in the curvature temporary storage unit; and registering a curvature sequence registered in the upward curvature temporary storage unit in the curvature curve pattern dictionary. a curvature curve error calculation unit that registers in the curvature error temporary storage unit a minimum curve error value when approximating each curvature curve pattern and an assumed center coordinate value corresponding to the minimum curve error value; A fingerprint center characterized by having a center coordinate detection unit that outputs a minimum value among minimum curve error values of each curvature curve pattern registered in a temporary storage unit and assumed center coordinates corresponding to the minimum value. Detection method.
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP63099335A JP2690103B2 (en) | 1988-04-23 | 1988-04-23 | Fingerprint center detection device |
| EP89107302A EP0339527B1 (en) | 1988-04-23 | 1989-04-21 | Fingerprint processing system capable of detecting a core of a fingerprint image by curvature parameters |
| DE68928154T DE68928154T2 (en) | 1988-04-23 | 1989-04-21 | Fingerprint processing system, suitable for determining the core of a fingerprint image by means of curvature parameters |
| US07/342,047 US5040224A (en) | 1988-04-23 | 1989-04-24 | Fingerprint processing system capable of detecting a core of a fingerprint image by statistically processing parameters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP63099335A JP2690103B2 (en) | 1988-04-23 | 1988-04-23 | Fingerprint center detection device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPH01271884A true JPH01271884A (en) | 1989-10-30 |
| JP2690103B2 JP2690103B2 (en) | 1997-12-10 |
Family
ID=14244758
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP63099335A Expired - Lifetime JP2690103B2 (en) | 1988-04-23 | 1988-04-23 | Fingerprint center detection device |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP2690103B2 (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0773508A2 (en) | 1995-11-08 | 1997-05-14 | Nec Corporation | Apparatus for extracting fingerprint features |
| US5848176A (en) * | 1995-04-04 | 1998-12-08 | Nec Corporation | Fingerprint fingertip orientation detection method and device |
| US6067369A (en) * | 1996-12-16 | 2000-05-23 | Nec Corporation | Image feature extractor and an image feature analyzer |
| JP2007226746A (en) * | 2006-02-27 | 2007-09-06 | Nec Corp | Fingerprint matching device, fingerprint pattern area extraction device, quality determination device, and method and program therefor |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3827567B2 (en) | 2001-12-05 | 2006-09-27 | 日本電気株式会社 | Fingerprint verification method and apparatus |
| JP4941791B2 (en) | 2007-04-23 | 2012-05-30 | 日本電気株式会社 | Two-dimensional pattern matching method, feature extraction method, apparatus and program used therefor |
-
1988
- 1988-04-23 JP JP63099335A patent/JP2690103B2/en not_active Expired - Lifetime
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5848176A (en) * | 1995-04-04 | 1998-12-08 | Nec Corporation | Fingerprint fingertip orientation detection method and device |
| EP0773508A2 (en) | 1995-11-08 | 1997-05-14 | Nec Corporation | Apparatus for extracting fingerprint features |
| US5832102A (en) * | 1995-11-08 | 1998-11-03 | Nec Corporation | Apparatus for extracting fingerprint features |
| US6067369A (en) * | 1996-12-16 | 2000-05-23 | Nec Corporation | Image feature extractor and an image feature analyzer |
| US6243492B1 (en) | 1996-12-16 | 2001-06-05 | Nec Corporation | Image feature extractor, an image feature analyzer and an image matching system |
| JP2007226746A (en) * | 2006-02-27 | 2007-09-06 | Nec Corp | Fingerprint matching device, fingerprint pattern area extraction device, quality determination device, and method and program therefor |
| US7885437B2 (en) | 2006-02-27 | 2011-02-08 | Nec Corporation | Fingerprint collation apparatus, fingerprint pattern area extracting apparatus and quality judging apparatus, and method and program of the same |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2690103B2 (en) | 1997-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP0470530B1 (en) | Fingerprint verification method | |
| JP2776757B2 (en) | Fingerprint finger axis detection device | |
| JP3742279B2 (en) | Image collation apparatus, image collation method, and recording medium recording image collation program | |
| EP0128820A2 (en) | Pattern matching method and apparatus | |
| CN110926330B (en) | Image processing apparatus, image processing method, and program | |
| JP4901676B2 (en) | License plate information processing apparatus and license plate information processing method | |
| AU722594B2 (en) | Apparatus for extracting fingerprint features | |
| US8005262B2 (en) | System and method for video object identification | |
| US5887083A (en) | Method of processing image information based on object model | |
| JP2011221699A (en) | Operation instruction recognition device and robot | |
| JPH01271884A (en) | Detecting system for center of fingerprint | |
| JPS6310472B2 (en) | ||
| JP2866461B2 (en) | Fingerprint collation device | |
| JPH01271883A (en) | Detecting system for center of fingerprint | |
| JP3574562B2 (en) | Method for recognizing and detecting deformation of a series of patterns in an image and recording medium storing the program | |
| JP2734245B2 (en) | Image feature extraction method and apparatus | |
| JP2002536709A (en) | Method and apparatus for calculating a similarity between a first structure and at least one preset second structure | |
| JP2702307B2 (en) | Fingerprint correction system | |
| JP2001076142A (en) | Method and device for collating fingerprint image and recording medium with this method recorded thereon | |
| JP3700675B2 (en) | Road white line recognition device | |
| JP3977473B2 (en) | Handwritten character recognition method and handwritten character recognition apparatus | |
| JP2008502955A (en) | Fingerprint authentication method including movement of control point | |
| CN113221891B (en) | Method and device for adjusting identification angle of palm vein image | |
| JP3718105B2 (en) | Form reading device, form reading method, and storage medium | |
| CN112348105B (en) | Unmanned aerial vehicle image matching optimization method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| R250 | Receipt of annual fees |
Free format text: JAPANESE INTERMEDIATE CODE: R250 |
|
| S531 | Written request for registration of change of domicile |
Free format text: JAPANESE INTERMEDIATE CODE: R313531 |
|
| R350 | Written notification of registration of transfer |
Free format text: JAPANESE INTERMEDIATE CODE: R350 |
|
| R250 | Receipt of annual fees |
Free format text: JAPANESE INTERMEDIATE CODE: R250 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20070829 Year of fee payment: 10 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20080829 Year of fee payment: 11 |
|
| EXPY | Cancellation because of completion of term | ||
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20080829 Year of fee payment: 11 |