JP4248892B2 - Fingerprint verification device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an apparatus for collating input fingerprint image data with registered fingerprint image data.
[0002]
[Prior art]
In a conventional fingerprint collation device, a gray level value of a pixel at the same position on an image captured after a predetermined time and a predetermined time is detected and compared, and forgery input is determined based on the result. (For example, refer to Patent Document 1).
[0003]
[Patent Document 1]
Japanese Patent Laid-Open No. 61-255481 (Page 499 (Example) and FIG. 4)
[0004]
[Problems to be solved by the invention]
In the conventional fingerprint collation device as described above, it is determined whether the fingerprint image is a fingerprint or a counterfeit fingerprint based on a change in the temporal density value of fingerprint image data (due to sweating action). There is a problem that it may be determined that the fingerprint is used.
[0005]
The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a fingerprint collation apparatus that can make a correct determination even for a person with less perspiration.
[0006]
[Means for Solving the Problems]
A fingerprint collation device according to the present invention is a device for collating input fingerprint image data with pre-registered fingerprint image data to determine the degree of coincidence between them, and is a part of the input fingerprint image data Gray value calculation means for calculating the temporal change in the movement amount of the feature point data, and when the temporal change in the gray value is equal to or less than a specified value, the temporal movement of the input fingerprint image data Fingerprint movement amount calculation means for calculating a change, and if the temporal change of the gray value is larger than a prescribed value, it is determined that the input fingerprint image data is from a human finger, and the temporal movement of the movement amount Do if variation is not less than the predetermined value, the fingerprint image data that is the input is judged to be due to a human finger, if the temporal variation is less than the predetermined value of the moving amount, the fingerprint image data that has been said input By forged fingerprint It is obtained by a counterfeit fingerprint determining means for determining as.
[0007]
DETAILED DESCRIPTION OF THE INVENTION
Embodiment 1 FIG.
FIG. 1 and FIG. 2 are diagrams showing an embodiment of the first, second and fifth inventions of the present invention. FIG. 1 is an overall configuration diagram and FIG. 2 is an operation flowchart.
In FIG. 1, a fingerprint sensor 1 includes a light source, a prism, an image sensor, and the like. The fingerprint sensor 1 captures a fingerprint image of a finger placed at a predetermined position and outputs fingerprint image data. The image memory 2 stores fingerprint image data from the fingerprint sensor 1.
[0008]
The image processing means 3 extracts feature point data necessary for collation from the fingerprint image data in the image memory 2. The feature point data storage unit 4 stores feature point data of fingerprint image data registered in advance. The collation processing means 5 collates and calculates the feature point data of the fingerprint image data extracted by the image processing means 3 with the feature point data of the registered fingerprint image data stored in the feature point data storage unit 4 so as to match them. Determine the degree.
[0009]
The result display unit 6 displays the collation result by the collation processing means 5. The forged fingerprint determination means 7 includes a gray value calculation means 7A and a fingerprint movement amount calculation means 7B. The fingerprint image data from the image memory 2 or the feature point data from the image processing means 3 is input, and the input fingerprint image is a forgery fingerprint. Determine if there is. The fingerprint device power-off means 8 turns off the power of the fingerprint collation measure when the forged fingerprint determination means 7 determines that the fingerprint is forged.
[0010]
In the fingerprint collation measure configured as described above, fingerprint image data output from the fingerprint sensor 1 is stored in the image memory 2, and feature point data is extracted by the image processing means 3. The collation processing unit 5 collates the feature point data from the image processing unit 3 with the feature point data registered in the feature point data storage unit 4, determines the degree of coincidence between them, and displays the result on the result display unit 6. indicate.
[0011]
On the other hand, the counterfeit fingerprint determination means 7 determines whether the input fingerprint is a human finger or a counterfeit fingerprint, and if it is counterfeit, displays it on the result display unit 6 and warns that it is a counterfeit fingerprint. . In addition, when the number of times determined to be forgery reaches a predetermined number, the fingerprint device power-off device 8 turns off the power of the fingerprint collation device to improve security.
[0012]
Next, the forged fingerprint determination operation according to this embodiment will be described with reference to FIG. In step S1, fingerprint image data from the image memory 2 (feature point data of the image processing means 3 may be used) is acquired. In step S2, the gray value calculator 7A calculates the gray value G (t) (t is time) of the pixels of the fingerprint image data. In step S3, it is determined whether time t is zero. If it is zero, the process returns to step S1, and if it is not zero, the process proceeds to step S4.
[0013]
In step S4, the absolute value of the difference between the gray value G (t = i) when t = i (i is an integer) and the gray value G (t = i + 1) when t = i + 1 is the predetermined value Tg. Determine whether: That is, if the difference between the shade values is larger than the predetermined value Tg, it is a change in the shade value due to the sweating action, so that it is determined to be due to a living body, and the process proceeds to step S5 to determine the fingerprint.
On the other hand, if the difference between the gray values is equal to or smaller than the predetermined value Tg, the process proceeds to step S6, where t = i and the fingerprint movement amount M (t = i + 1) are calculated from the fingerprint image data of t = i + 1.
[0014]
The movement amount M can be obtained by image pattern matching, which is a well-known technique. In step S7, it is determined whether the movement amount M (t) is equal to or greater than a predetermined value Tm. If the movement amount M (t) is greater than or equal to the predetermined value Tm, it is determined that the finger is moving (due to rotation, twisting, etc.), and the fingerprint is determined in step S5. However, if the movement amount M (t) is less than the predetermined value, it is determined that the phenomenon does not occur due to a non-living body, and it is determined as a forged fingerprint in step S8.
Here, steps S1 to S4 constitute a gray value calculation means 7A, steps S6 and S7 constitute a fingerprint movement amount calculation means 7B, and steps S5 and S8 constitute a counterfeit fingerprint determination means 7.
In this way, it is possible to correctly determine a person who has been determined to be a fake finger due to low sweating or the like.
[0015]
Embodiment 2. FIG.
FIG. 3 is an explanatory diagram of direction angle data showing an embodiment of the third invention of the present invention. 1 and 2 are shared by the second embodiment.
In this embodiment, the fingerprint image data itself in the image memory 2 is used for calculating the amount of fingerprint movement in the first embodiment, whereas the fingerprint image data output from the image processing means 3 is subjected to image processing. The direction angle data 11 is used. As shown in FIG. 3, the direction angle data 11 is global fingerprint data indicating the vortex of the fingerprint and the flow of the fingerprint.
In this way, by using the direction angle data 11, it is possible to shorten the time for calculating the amount of movement of the fingerprint compared to the case of the first embodiment.
[0017]
Embodiment 3 FIG.
FIG. 4 is an explanatory diagram of feature point data showing an embodiment of the fourth invention of the present invention. 1 and 2 are also used in the third embodiment.
In this embodiment, the fingerprint image data itself in the image memory 2 is used for calculating the amount of fingerprint movement in the first embodiment, whereas the fingerprint image data output from the image processing means 3 is subjected to image processing. The feature point data 12 is used. As shown in FIG. 4, the feature point data 11 is fingerprint data representing the positions and angles of the fingerprint end points 12a and branch points 12b.
[0018]
In this way, by using the feature point data 12, it is possible to shorten the time for calculating the amount of fingerprint movement compared to the case of the first embodiment.
[0019]
【The invention's effect】
As described above, according to the present invention, in the apparatus for collating input fingerprint image data with previously registered fingerprint image data and determining the degree of coincidence between them , the feature is a part of the input fingerprint image data Gradation value calculation means for calculating a temporal change in the movement amount of the point data, and a temporal change in the movement amount of the input fingerprint image data when the temporal change in the gray value is equal to or less than a specified value. If the temporal change in the gray value is greater than a prescribed value, it is determined that the input fingerprint image data is from a human finger, and the temporal movement of the movement amount is calculated. if the change is greater than a predetermined value, the fingerprint image data that is the input is judged to be due to a human finger, if the temporal variation of the amount of movement is less than the predetermined value, the fingerprint image data that is the input By counterfeit fingerprint Since a and determining counterfeit fingerprint determination unit thing, people who have been determined to be a false finger for sweating is small or the like can be determined correctly.
[Brief description of the drawings]
FIG. 1 is an overall configuration diagram showing Embodiment 1 of the present invention.
FIG. 2 is an operation flowchart showing Embodiment 1 of the present invention.
FIG. 3 is an explanatory diagram of direction angle data showing a second embodiment of the present invention.
FIG. 4 is an explanatory diagram of feature point data showing Embodiment 3 of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Fingerprint sensor, 2 Image memory, 3 Image processing means, 4 Feature point data storage part, 5 Collation processing means, 7 Forged fingerprint determination means, 7A Gray value calculation means, 7B Fingerprint movement amount calculation means, 8 Fingerprint device power-off device 11 Direction angle data, 12 Feature point data.
S1 to S4 Gray value calculation means, S6, S7 Fingerprint movement amount calculation means, 7 Counterfeit fingerprint determination means.

Claims (2)

In apparatus for determining both the degree of matching by matching the input fingerprint image data with previously registered fingerprint image data,
A gray value calculation means for calculating a temporal change in the gray value of the feature point data that is a part of the input fingerprint image data;
A fingerprint movement amount calculation means for calculating a temporal change in the movement amount of the inputted fingerprint image data when the temporal change of the gray value is not more than a specified value ;
If the temporal change of the gray value is larger than the specified value , the input fingerprint image data is determined to be from a human finger,
If the temporal change in the movement amount is equal to or greater than a predetermined value, it is determined that the input fingerprint image data is from a human finger,
A fingerprint collation apparatus comprising: a counterfeit fingerprint determination unit that determines that the input fingerprint image data is based on a counterfeit fingerprint if the temporal change in the movement amount is less than a predetermined value. False fingerprints and the determined number of fingerprint identification apparatus according to claim 1 Symbol mounting, characterized in that a fingerprint device power-off device for turning off the power supply reaches a predetermined number of times.

Images