JP5477906B2 - Fingerprint feature point type determination device, fingerprint feature point type determination method, fingerprint feature point type determination program - Google Patents

Description

  The present invention relates to a system for performing fingerprint collation using a fingerprint image, and more particularly to a system for setting attribute information of each feature point in a fingerprint image.

  In a fingerprint feature point information setting system for setting information related to feature points in an input fingerprint image, generally, as shown in FIG. 1, the inside of each fingerprint image is based on the fingerprint ridge shape of the input fingerprint image. In addition to designating the position of the feature point, the direction as the attribute of the feature point (hereinafter referred to as “feature point direction”) and the type of each feature point are set for each feature point.

  Here, as shown in FIG. 2, there are branch points and end points as types of feature points, and feature point information including the position of the feature point in each fingerprint image, the direction of each feature point, and the type is determined. By performing collation based on this feature point information, it is possible to speed up collation processing and increase collation accuracy.

  By the way, in the setting of feature point information (feature point position designation, feature point direction designation, and feature point type setting), the input fingerprint image is artificially set by, for example, an operator of a fingerprint collation device. Often.

For this reason, inconveniences such as oversight and oversight of feature points in the fingerprint image by the operator of the fingerprint collation apparatus, and further input errors of feature point information may occur.
In addition, some of the fingerprint images that have been input, such as vestigial fingerprints, have ridges in the image that are unclear and of low quality. For example, feature points can be located in the image. Even in such a case, when it is difficult to determine the type of each feature point, there is a disadvantage that it takes time to determine the type of the feature point or a mistake in determining the type occurs.
On the other hand, when there are a large number of input fingerprint images or when many feature points included in the fingerprint image are included, it is required to quickly and accurately determine the type of each feature point. .

  As a related technology, fingerprint data including information about feature points is created for each fingerprint image, and a common feature point that is the same feature point is obtained from the correspondence of feature points among a plurality of fingerprint data. A technique for improving collation performance by creating synthetic fingerprint data based on common feature points and performing fingerprint collation using the synthetic fingerprint data is disclosed (Patent Document 1).

JP 2001-344604 A

  However, even in the related technique disclosed in Patent Document 1, when a large amount of fingerprint images are input or when many feature points are included in the fingerprint image, the feature point information is set in an enormous amount. There is an inconvenience that it takes time and effort.

[Object of invention]
The present invention improves the inconveniences of the related art and enables a fingerprint feature point type determination system, a fingerprint feature point type determination method, and a fingerprint feature point that can quickly set the type of each feature point set in a fingerprint image. The purpose is to provide a type determination program.

  In order to achieve the above object, the fingerprint feature point type determination device according to the present invention is configured so that the position of the feature point included in the fingerprint image and the feature point direction at each feature point are set in advance. A fingerprint feature point type determining device including a feature point type determining unit for determining a type of feature point, wherein the feature point type determining unit includes a length parallel to the feature point direction and a plane in contact with the feature point. A rectangular area A having a predetermined size and having a predetermined size and a rectangular area B that is moved through the rectangular area A with respect to an orthogonal line segment that passes through the feature point and is orthogonal to the feature point direction An area setting means, an in-rectangular average density calculation means for calculating average density values a and b in the rectangles, which are average density values in the set rectangular areas A and B, and the in-rectangular average density value a Average density value in rectangle A feature point type determination unit that determines that the feature point type is a branch point when larger than the average density value a within the rectangle and is smaller than the average density value b within the rectangle. It is characterized by having.

  The fingerprint feature point type determining method according to the present invention determines the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset. A fingerprint feature point type determination method for setting a rectangular area A having a preset size having a length parallel to the feature point direction and a plane in contact with the feature point, and passing through the feature point A rectangular area B which is the target moved in the rectangular area A with respect to an orthogonal line segment orthogonal to the feature point direction is set, and an average density value a in the rectangle which is an average density value in each of the set rectangular areas A and B And b are calculated, and when the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and the average density value a within the rectangle is determined as the average density within the rectangle If the value b is smaller than It is characterized by determining a type of point and end point.

  The fingerprint feature point type determination program according to the present invention determines the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset. A fingerprint feature point type determination program for setting a rectangular area A having a preset size having a length parallel to the feature point direction and a plane in contact with the feature point, and the feature point A rectangular area setting function for setting a rectangular area B that is moved in the rectangular area A with respect to an orthogonal line segment orthogonal to the feature point direction, and an average density value in each of the set rectangular areas A and B Intra-rectangular density value calculation function for calculating a certain average density value a and b in the rectangle, and when the average density value a in the rectangle is larger than the average density value b in the rectangle, the type of the feature point is determined as a branch point. , Serial rectangular average density value a is characterized by executing the endpoints and determine feature point type determination function a type of the feature point, to the computer when the rectangular average density value smaller than b.

  Since the present invention is configured and functions as described above, according to this, the feature point passes through the rectangular area A composed of the long side parallel to the preset feature point direction and the short side in contact with the feature point and the feature point. Rectangle area setting means for setting a rectangular area B that has been moved relative to the rectangular area A with respect to an orthogonal line segment that is orthogonal to the direction, and an average density calculation means for calculating an average density value in each of the rectangular areas A and B And the feature point type determining means for comparing the calculated average density value in the rectangle and determining the type of the feature point from the magnitude result thereof, thereby quickly determining the type of the feature point in the fingerprint image. It is possible to provide a fingerprint feature point type determination system, a fingerprint feature point type determination method, and a fingerprint feature point type determination program that can be performed.

It is explanatory drawing which shows an example of the position of the feature point in a fingerprint image, and a feature point direction. It is explanatory drawing which shows an example of the branch point and end point as a kind of feature point in a fingerprint image. It is explanatory drawing which shows the rectangular area | region which the rectangular area | region setting means in the fingerprint feature point classification determination system disclosed in FIG. 1 set on the fingerprint image. It is explanatory drawing which shows the method in which the rectangular area setting means in the fingerprint feature point type determination system disclosed in FIG. 1 sets a rectangular area. It is a schematic block diagram which shows one Embodiment in the fingerprint feature point classification determination system by this invention. It is a schematic block diagram which shows the internal structure of the feature point classification determination part in the fingerprint feature point classification determination system by this invention.

[Embodiment]
Next, the basic configuration content of the embodiment of the present invention will be described.

  As shown in FIG. 5, the fingerprint feature point information setting device 10 according to the present embodiment receives a fingerprint image input from the outside, and based on the ridge shape in the fingerprint image, the feature points in the fingerprint image. A feature point position specifying unit 1 that specifies the position of the feature point, a feature point direction determining unit 2 that determines a feature point direction at each specified feature point, and a type of each feature point for which the feature point direction has been determined. That is, it has a configuration including a feature point type determination unit (corresponding to a fingerprint feature point type determination device) 3 that determines whether each feature point is an end point or a branch point.

  When the feature point type determination unit 3 sets the position of the feature point in the input fingerprint image and the feature point direction at each feature point by the feature point position specification unit 1 and the feature point direction determination unit 2, respectively. Next, the type of each feature point is determined.

  The feature point type determination unit 3 is a rectangular area A having a preset size, which includes a long side (long) parallel to a preset feature point direction and a short side (flat) in contact with the feature point. , The feature point direction of this feature point and the orthogonal line segment (H) orthogonal to the feature point are set, and the rectangular region B that is the target moved from the rectangular region A to the orthogonal line segment (H) Rectangle area setting means 31 to be set; Rectangle average density calculation means 32 for calculating the average density values a and b in the rectangle as the average density values in the set rectangular areas A and B; and the average density in the rectangle A comparison is made between the value a and the average density value b within the rectangle, and based on the determination result (large or small), the feature point type determining means 33 determines whether the feature point type of this feature point is a branch point or an end point. It has.

Here, when the average density value a within the rectangle is larger than the average density value b within the rectangle (a> b), the feature point type determination unit 33 determines the type of the feature point as a branch point, and determines the average density within the rectangle. When the value a is smaller than the in-rectangular average density value b (a <b), the feature point type is determined as an end point.
In the present embodiment, the density value of each pixel in the fingerprint image is set in the range of 0 (black) to 255 (white).

In addition, the rectangular area setting unit 31 sets a rectangular area A composed of a long side parallel to the feature point direction and a short side in contact with the feature point, and then moves the rectangular area A with respect to the feature point. A region moved symmetrically) may be set as the rectangular region B.
Thereby, the rectangular areas A and B can be set more quickly without setting the orthogonal line segment (H) at each feature point.
It is assumed that the rectangular areas A and B set corresponding to each feature point have the same area. Further, the shape of the set region (here, the rectangular regions A and B) may be a shape other than a rectangle.

The rectangular areas A and B set by the rectangular area setting unit 31 (long side and short side length) are assumed to be equal to each other. In addition, the long sides and short sides of the rectangular areas A and B are set in advance.
Note that the length of the short sides of the rectangular areas A and B may be set to the average width value of the fingerprint ridges included in the input fingerprint image.

  The in-rectangular average density calculation means 32 calculates the average density value of the pixels in the rectangular area A by calculating the average density value of the pixels included in each of the rectangular areas A and B set by the rectangular area setting means 31. The average density value “a” within the rectangle and the average density value “b” within the rectangle which are the average density values of the pixels in the rectangular area B are calculated.

  As described above, the feature point type determination unit 33 compares the sizes of the average density value a within the rectangle and the average density value b within the rectangle calculated by the average density calculation unit 32 within the rectangle, and calculates the average density value a within the rectangle a. Is larger than the average density value b in the rectangle (a> b), the type of the feature point is determined as a branch point, and the average density value a in the rectangle is smaller than the average density value b in the rectangle (a < In b), the type of this feature point is determined as an end point.

The feature point position specifying unit 1 in the present embodiment (fingerprint feature point information setting device 10) specifies the position of the feature point by plotting the position of the feature point in the input fingerprint image on the fingerprint image. .
In addition, the feature point direction determination unit 2 in the present embodiment determines the feature point direction for the feature points whose positions in the image are specified by the feature point position specification unit 1 based on the average pixel density value around each feature point. To do.
Here, the feature point direction determination unit 2 may be set to determine the feature point direction at each feature point by the method disclosed in Japanese Patent Application No. 2009-116893, for example.

  As described above, the feature point type determination unit 3 of the fingerprint feature point information setting device 10 determines the type of each feature point when the position and direction (feature point direction) of the feature point in the fingerprint image is set in advance. Can be determined quickly and accurately. For this reason, the fingerprint feature point information setting device 10 according to the present embodiment can quickly set information (feature point information: position, direction, type) related to feature points in each inputted fingerprint image. .

[Description of Operation of Embodiment]
Next, an outline of the operation of the first embodiment will be described.
First, the rectangular area setting unit 31 sets a rectangular area A having a predetermined size composed of a long side parallel to the feature point direction of a predetermined feature point and a short side in contact with the feature point, and A rectangular area B is set by moving the rectangular area A with respect to an orthogonal line segment that passes through the characteristic points and is orthogonal to the direction of the characteristic points (rectangular area setting step).
Next, the in-rectangular density value calculating means 32 calculates the in-rectangular average density values a and b which are average density values in the set rectangular areas A and B (intra-rectangular density value calculation process), and finally. The feature point type determination means 33 determines that the feature point type is a branch point when the calculated average density value a within the rectangle is larger than the average density value b within the rectangle, and the average density value a within the rectangle is If the average density value in the rectangle is smaller than b, the feature point type is determined as an end point (feature point type determination step).
Here, the execution contents of the rectangular area setting step, the in-rectangular density value calculation step, and the feature point type determination step may be programmed and executed by a computer.

  Next, the operation of the fingerprint feature point information setting apparatus according to the present embodiment will be described based on the block diagrams of FIGS.

  First, the rectangular area setting unit 31 obtains a fingerprint image in which a feature point position (feature point position) and a feature point direction are set, and each feature point in the fingerprint image is parallel to the feature point direction of the feature point. A rectangular area A composed of a long side and a short side that is in contact with the feature point and orthogonal to the feature point direction is set.

  Further, as shown in FIG. 4, the rectangular area setting unit 31 sets an orthogonal line segment (H) orthogonal to the feature point direction (that is, parallel to the set short side) and the rectangle. A rectangular area that is subject to the area A with respect to the orthogonal line segment (H) is set as a rectangular area B (rectangular area setting step).

  Note that the rectangular area setting unit 31 may set the rectangular area A as a rectangular area B after setting the rectangular area A and then subjecting the rectangular area A to the characteristic points (point-symmetrical movement).

  Next, the intra-rectangular density value calculation means 32, when the rectangular areas A and B for each feature point are set, the intra-rectangular average that is the average density value of the pixels included in the rectangular areas A and B, respectively. The density values a and b are respectively calculated (intra-rectangular density value calculation process). In the present embodiment, as described above, the density value of each pixel in the fingerprint image is set in the range of 0 (black) to 255 (white).

Next, the feature point type determination unit 33 compares the calculated average density value “a” within the rectangle and the average density value “b” within the rectangle.
Here, when the average density value a within the rectangle is larger than the average density value b within the rectangle, the feature point type determination unit 33 determines the feature point type as a branch point, and the average density value a within the rectangle is within the rectangle. If it is smaller than the average density value b, the corresponding feature point type is determined as an end point (feature point type determination step).

  As described above, in the feature point type determination unit 3 of the fingerprint feature point information setting system (FIG. 4) according to the present embodiment, the position and the feature point direction of the fingerprint feature point designated by the operator in the fingerprint image are determined. In this case, the type of each feature point can be determined quickly and accurately. In this way, in the residual fingerprint input process for setting the feature point information for fingerprint verification on the input fingerprint image, It is possible to effectively reduce the labor for inputting and setting feature point information.

About the above-mentioned embodiment, it is as follows when the summary of the novel technical content is put together.
In addition, although one part or all part of the said embodiment is put together as follows as a novel technique, this invention is not necessarily limited to this.

(Appendix 1)
Fingerprint feature point type determination provided with a feature point type determination unit that determines the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset. A device,
The feature point type determination unit
A rectangular area A having a preset size having a length parallel to the feature point direction and a flat surface in contact with the feature point is set, and an orthogonal line segment that passes through the feature point and is orthogonal to the feature point direction is set. A rectangular area setting means for setting a rectangular area B which is the target moved to the rectangular area A;
In-rectangular average density calculation means for calculating average density values a and b in the rectangle, which are average density values in the set rectangular areas A and B, respectively.
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination device comprising: a feature point type determination unit that determines the type of the feature point as an end point.

(Appendix 2)
Fingerprint feature point type determination device including a feature point type determination unit that determines the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are set Because
The feature point type determination unit
A rectangular area A having a preset size having a long side parallel to the feature point direction and a short side in contact with the feature point is set, and the rectangular area A is moved relative to the feature point A rectangular area setting means for setting an area as the rectangular area B;
In-rectangular average density calculation means for calculating average density values a and b in the rectangle, which are average density values in the set rectangular areas A and B, respectively.
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination device comprising: a feature point type determination unit that determines the type of the feature point as an end point.

(Appendix 3)
In the fingerprint feature point type determination device according to attachment 1 or 2,
The rectangular area setting means includes:
An advisory feature point type determination apparatus characterized in that the flat length of the rectangular area A is set as an average width value of fingerprint ridges included in the fingerprint image.

(Appendix 4)
A fingerprint feature point type determination method for determining the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset,
A rectangular area A having a preset size having a length parallel to the feature point direction and a flat surface in contact with the feature point;
Set a rectangular area B that has moved the rectangular area A with respect to an orthogonal line segment that passes through the feature point and is orthogonal to the feature point direction;
In-rectangular average density values a and b, which are average density values in the set rectangular area A and B, respectively, are calculated,
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination method, wherein the feature point type is determined as an end point.

(Appendix 5)
A fingerprint feature point type determination method for determining the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset,
A rectangular area A having a preset size having a length parallel to the feature point direction and a flat surface in contact with the feature point;
Set a rectangular area B that is the target of moving the rectangular area A with respect to the feature point;
In-rectangular average density values a and b, which are average density values in the set rectangular area A and B, respectively, are calculated,
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination method, wherein the feature point type is determined as an end point.

(Appendix 6)
A fingerprint feature point type determination program for determining the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset,
A rectangular area A having a preset size having a length parallel to the feature point direction and a flat surface in contact with the feature point is set, and an orthogonal line segment that passes through the feature point and is orthogonal to the feature point direction is set. A rectangular area setting function for setting a rectangular area B that has been moved to the rectangular area A;
Intra-rectangular density value calculation function for calculating average density values a and b in the rectangle, which are average density values in the set rectangular areas A and B, respectively.
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination program for causing a computer to execute a feature point type determination function for determining a feature point type as an end point.

(Appendix 7)
A fingerprint feature point type determination program for determining the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset,
A rectangular area A having a preset size having a length parallel to the feature point direction and a plane in contact with the feature point is set, and an area in which the rectangular area A is subject to the feature point is moved A rectangular area setting function for setting the rectangular area B;
Intra-rectangular density value calculation function for calculating average density values a and b in the rectangle, which are average density values in the set rectangular areas A and B, respectively.
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination program for causing a computer to execute a feature point type determination function for determining a feature point type as an end point.

  The present invention can be effectively applied to a fingerprint feature point information registration system that registers information related to feature points in a fingerprint image for fingerprint verification.

DESCRIPTION OF SYMBOLS 1 Feature point position designation | designated part 2 Feature point direction determination part 3 Feature point type determination part 10 Fingerprint feature point information setting apparatus 31 Rectangular area setting means 32 Rectangle average density calculation means 33 Feature point type determination means

Claims (7)

Fingerprint feature point type determination provided with a feature point type determination unit that determines the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset. A device,
The feature point type determination unit
A rectangular area A having a preset size having a length parallel to the feature point direction and a flat surface in contact with the feature point is set, and an orthogonal line segment that passes through the feature point and is orthogonal to the feature point direction is set. A rectangular area setting means for setting a rectangular area B which is the target moved to the rectangular area A;
In-rectangular average density calculation means for calculating average density values a and b in the rectangle, which are average density values in the set rectangular areas A and B, respectively.
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination device comprising: a feature point type determination unit that determines the type of the feature point as an end point. Fingerprint feature point type determination device including a feature point type determination unit that determines the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are set Because
The feature point type determination unit
A rectangular area A having a preset size having a long side parallel to the feature point direction and a short side in contact with the feature point is set, and the rectangular area A is moved relative to the feature point A rectangular area setting means for setting an area as the rectangular area B;
In-rectangular average density calculation means for calculating average density values a and b in the rectangle, which are average density values in the set rectangular areas A and B, respectively.
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination device comprising: a feature point type determination unit that determines the type of the feature point as an end point. In the fingerprint feature point type determination device according to claim 1 or 2,
The rectangular area setting means includes:
An advisory feature point type determination apparatus characterized in that the flat length of the rectangular area A is set as an average width value of fingerprint ridges included in the fingerprint image. A fingerprint feature point type determination method for determining the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset,
A rectangular area A having a preset size having a length parallel to the feature point direction and a flat surface in contact with the feature point;
Set a rectangular area B that has moved the rectangular area A with respect to an orthogonal line segment that passes through the feature point and is orthogonal to the feature point direction;
In-rectangular average density values a and b, which are average density values in the set rectangular area A and B, respectively, are calculated,
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination method, wherein the feature point type is determined as an end point. A fingerprint feature point type determination method for determining the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset,
A rectangular area A having a preset size having a length parallel to the feature point direction and a flat surface in contact with the feature point;
Set a rectangular area B that is the target of moving the rectangular area A with respect to the feature point;
In-rectangular average density values a and b, which are average density values in the set rectangular area A and B, respectively, are calculated,
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination method, wherein the feature point type is determined as an end point. A fingerprint feature point type determination program for determining the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset,
A rectangular area A having a preset size having a length parallel to the feature point direction and a flat surface in contact with the feature point is set, and an orthogonal line segment that passes through the feature point and is orthogonal to the feature point direction is set. A rectangular area setting function for setting a rectangular area B that has been moved to the rectangular area A;
Intra-rectangular density value calculation function for calculating average density values a and b in the rectangle, which are average density values in the set rectangular areas A and B, respectively.
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination program for causing a computer to execute a feature point type determination function for determining a feature point type as an end point. A fingerprint feature point type determination program for determining the type of each feature point when the position of the feature point included in the fingerprint image and the feature point direction at each feature point are preset,
A rectangular area A having a preset size having a length parallel to the feature point direction and a plane in contact with the feature point is set, and an area in which the rectangular area A is subject to the feature point is moved A rectangular area setting function for setting the rectangular area B;
Intra-rectangular density value calculation function for calculating average density values a and b in the rectangle, which are average density values in the set rectangular areas A and B, respectively.
When the average density value a within the rectangle is larger than the average density value b within the rectangle, the type of the feature point is determined as a branch point, and when the average density value a within the rectangle is smaller than the average density value b within the rectangle. A fingerprint feature point type determination program for causing a computer to execute a feature point type determination function for determining a feature point type as an end point.

Images