JPS63138477A - Control system for fingerprint data flow by fingerprint quality - Google Patents

Abstract

PURPOSE:To reduce fingerprint data capacity and to shorten data transfer time by controlling fingerprint data flow based on fingerprint quality. CONSTITUTION:The display of a core line or an original picture is controlled by means of the information on the fingerprint quality. When the quality of a fingerprint is decided 'excellent', for example, by a fingerprint reading device 2, only the feature point list of the fingerprint is stored in a feature point file 8 via an original picture file 4a and a core line file 6. The feature point list is stored in the file 4a with the fingerprint quality 'failure' and the fingerprint data is erased. In such a way, the fingerprint data flow is controlled based on the quality of fingerprints. Thus the capacity of the file 4a and the time needed for operator's correcting jobs via a fingerprint monitor device 3 is also shortened.

Description

[Detailed Description of the Invention] [Industrial Field of Application] The present invention is a method for automatically verifying fingerprints using minutiae of fingerprints.
In a 4-turn recognition system, a control method that controls the flow of fingerprint image data and performs confirmation and correction operations on the fingerprint image data, in particular using fingerprint image quality information, regarding feature point extraction processing and feature point confirmation processing. Regarding.

[Conventional technology]

Conventionally, in a pattern recognition system that automatically verifies fingerprints, a CCD scanner is used to read the fingerprint to obtain image data having shading, and feature points are automatically extracted from this image data. Therefore, it is generally impossible to achieve a 100% extraction rate of features from fingerprints.

For this reason, the extraction contents of the feature points are checked using a manual machine (CRT, etc.), and as a result, the extraction contents are corrected if necessary. Generally, fingerprint data is displayed on a display in order to perform this confirmation and correction, but in this case, all image information read from the fingerprint reader is simply sent to the fingerprint reader and displayed on the display. The verification and correction of fingerprint data is performed entirely at the discretion of the operator.

[Problem that the invention seeks to solve]

By the way, when a fingerprint reading device and a plurality of fingerprint monitoring devices are provided, writing of fingerprint image data obtained by the fingerprint reading device to an original image file and displaying from the original image file to the plurality of fingerprint monitoring devices are executed asynchronously. In the case of a system equipped with a fingerprint monitor in consideration of the balance between the processing time of write processing and display processing, if all the fingerprint image information obtained from the fingerprint reader is displayed on the fingerprint monitor, it will take a long time to transfer the data. Moreover, there is a problem that the capacity of a file such as a disk that temporarily stores the fingerprint image becomes enormous.

On the other hand, as described above, in a pattern recognition system that automatically matches fingerprints, it is necessary to confirm and correct fingerprint image data in order to improve matching accuracy.

Therefore, when checking and correcting the fingerprint image information by displaying the entire image on the fingerprint monitor device.

The decision as to whether or not to modify the fingerprint image information is entirely made by the operator, and there is a problem in that the content of modification varies depending on the difference in fliff between operators, resulting in uneven quality of the fingerprint image information.

[Means for solving problems]

The fingerprint data flow control method based on fingerprint quality according to the present invention includes a fingerprint reading means for extracting a minutiae list including fingerprint quality information, a core line, and an original image from a fingerprint image, and inputting the core line or original image based on the fingerprint quality information. . fingerprint quality determining means for determining whether or not the feature point list is stored in a file means, and output means for outputting the core line or original image to the file means according to the determination by the fingerprint quality determining means; It has an image search means for searching, and a fingerprint monitor means for displaying the search contents retrieved by the image search means and for confirming and correcting the extracted contents.

The present invention is characterized in that display of the core line or the original image is controlled using the fingerprint quality information.

〔Example〕

The present invention will be explained below with reference to Examples.

Referring to FIG. 1, the central processing unit (CPU).

Fingerprint reader 2. Multiple fingerprint monitoring devices3. and a magnetic disk device 4 are connected. This central processing unit 1 has nine image processing units 1a, fingerprint quality determination unit lb, and image output unit 1.
e, and an image search section 1d. The fingerprint reading device 2 also includes a fingerprint quality extraction section 2a.

The fingerprint reading device 2 scans a fingerprint base paper on which, for example, 10 fingerprints of one person have been pressed with ink, as shown in FIG.
), the position coordinates M (x + y) of the minutiae points (end points and branch points) of the fingerprint for each finger are determined by the ridge direction d of the fingerprint, and the nearby minutiae points ( R1-
The number of ridges (r1 to r4) between R4) is extracted. Then, a reading number (processing number) is printed on the back side of the fingerprint base paper read by the fingerprint reading device 2.

The minutiae extracted by the fingerprint reader 2 and the number of ridges between neighboring minutiae points are used during fingerprint verification, and in particular, the number of ridges (relations) between neighboring minutiae points are verified as shown in FIG. 3(b). It is used as important information.

In addition, in the example shown in FIG. 3(b), in the case of fingerprint (1) and fingerprint (3), in the number of ridges between minutiae M1 and M2,
It is shown that each of the three fingerprints is equal, but in the case of fingerprint (2), one fingerprint is different from fingerprint (1).

Fingerprint information (fingerprint data) output from the fingerprint reader 2
is a feature point list as shown in FIG. 4(a).

These are the core line and the original drawing. The feature point list is as follows.

As shown in Figure 4, the recognition unit (I
cation)-Part: Is zone part [Z (Z
one ) Part ) * and detail part (M (M
inutiae) -Part)kara [, I-P
The fingerprint data reading number (individual descriptive information such as gender, date of birth, etc.), coordinate axes (Xo, Yc, Do), fingerprint quality, etc. are set in art, and the fingerprint image (32 Each area (0.8m x 0.8
Information on whether the image (question) is clear or unclear (expressed in 2 bits as shown in Figure 4(c), OO: unclear area, 01: unstable area.

10: clear area, 11: curved area) are set, and further,
M-Part has feature points (position coordinates x, y, ridge direction d
, neighboring feature points R4 to R4, number of ridges between neighboring feature points r1 to
ra) is set.

The core line is the skeleton part of the fingerprint ridge as shown in FIG. 4(d), and the 512
Each area of x640 mesh (0.05m x 0.05m)
(O: Not a core line. 1: A core line.) On the other hand, the original image represents the ink density of the fingerprint ridge pressed onto the fingerprint base paper as shown in Figure 4(f). Each area of the 512x640 mesh (0,
0.05 x 0.05 m) density in 4 bits.

It is expressed using 16 levels of gradation (Figure 4-)).

The fingerprint monitor device 3 has feature points, coordinate axes, and zones.
E) Display the core line and original image on the screen, and use the coordinate axes.

It is used as a man-machine device to correct ZONE and core wire. The operator uses the stylus (not shown) on the tablet part of the fingerprint monitor device 3 to
ZONE and skeleton lines can be corrected, and personal descriptive information and commands can be input using the keyboard section.Also, a hard copy of an image can be taken from the fingerprint monitor device 3 if necessary.

The magnetic disk device 4 is a file that temporarily stores the original image file.

The feature point list, skeleton lines, and original image are temporarily stored.

Here, the fingerprint quality extracted from the fingerprint original by the fingerprint quality extraction unit 2a of the fingerprint reading device 2 will be explained.

Fingerprint quality is defined by region quality and coordinate axis quality. Area quality expresses the sharpness of the fingerprint impression, and coordinate axis quality expresses the extraction accuracy of the coordinate axes.

(1) Area quality Area quality is the ratio of the number of clear areas to excellent, good, and fair.

It is evaluated on a four-point scale of not acceptable. The ratio of the number of sharp areas is determined for each section. Figure 5(a).

When determining the two plots with reference to (b) and (C),
Impression center information (Ie, J,) and impression edge information (Iz *
I, + J (+Jb), refer to the partition determination function shown in FIG. 5(b). At this time, first, Δt=l
I, -Itl/16.

Δr = l I c I r l / 16 pΔt
= ” c-Jt t/ t s e and Δb =
1 Jo-J, 1/16 are calculated, and based on these, the direct on the partition determination function is read to create 2 partitions A, B, and C.

Determine. That is, At, Bt, and Ct are found from Jt, and in the same way, Ar9Br#Cr, At#Bt, Ct, and Ab t B are obtained from Δr, Δt, and Δb, respectively.
b −C, is calculated.

Therefore, the A section is (At, A, , A, , A,),
B section is (BztBr-BtsB,) 7C section is (c
t - cr * ct , C,) (Fig. 5 (
c) ).

Once the 9 sections are determined as described above, the second section is determined.

Quality within each zone is required 0 Quality within zone Az e Bz
e and C are determined by the ratio of the number of sharp areas in each area in the Z-Part, and in this case, the following points are taken into consideration.

■There must be a clear area of more than a certain area within the section.

■There must be no blank areas within section B.

■There are many white areas in section C.

Then, the area quality is defined by the combination of the intra-section qualities A21 Bz+c.

(2) Coordinate axis quality The extraction of the coordinate axes consists of the extraction of the center point position and the extraction of the Y-axis stop direction, and the quality of the coordinate axes is evaluated in three stages: "excellent", "fair", and 'impossible' depending on the stability of the extraction. Ru.

With reference to FIGS. 6(a), (b), and (C),
Center point extraction will be explained. First, start point (shown as 81 to S5 in Fig. 6(, ), but here Sl
I will explain about it. ) to find the intersection point r with the -ridge ■. Then, the highest position H1 is determined by tracing the ridge line ■ corresponding to the intersection r, - in the left-right direction. Next, trace downward from the highest position H4, find the intersection r2 with the ridge ■, and do the same.

Highest position H2, intersection r3 z Find the highest position H3. Find the intersection r4 with the line ■. That is, H3→r4 is repeated. Then, the highest position H3 is defined as the center point.

Once the center point is defined, processing begins to extract the Y-axis stop direction.

From the center point obtained as described above, trace the ridge ■ where the center point exists equidistantly from the center point in the left and right direction,
As shown in FIG. 6(b), points t and r.

seek. A straight line connecting points t1 and r and two ridges on the outside ■
, ■ Find the intersections t2 and r2 + and t3 and r3.

Using these intersection points (t1 to t3 r r1 to rs) as a reference, trace each ridge at equal distances in the vertical direction and find the starting and ending points of the vector, and the direction vector L,
~L31 R1-R3 is determined. The length of the direction vector is made shorter as it goes further outside the ridge where the center point is located. As shown in FIG. 6(C), the direction of the sum of the busy direction vectors becomes the Y-axis stop direction, and is expressed as a Y-axis unit vector in FIG. 6(C).

The quality of the coordinate axes is determined using the following criteria in the center point and coordinate axes extraction process described above.

■ Coordinate axis quality 6 Excellent'' (a) When extracting the center point position, the center position must be found as the same position for all traces from three starting positions.

(c) The center point position must be found near the center of the fingerprint impression.

(b) Regarding the Y-axis pressure direction extraction, six vectors must be found with predetermined lengths.

b) In the Y-axis pressure direction extraction, the variance of the six vector directions must be below a certain value.

■ Coordinate axis quality is "fair" This is a case where the coordinate axes can be found even though all the conditions for "coordinate axis quality is excellent" are not satisfied.

■ Coordinate axis quality “Unavailable” Can't you find both the center point and the Y-axis stop direction?

Or there is a case where only one can be found.

(3) Fingerprint Quality Fingerprint quality is defined by a combination of region quality and coordinate axis quality.

Referring to FIG. 7, the fingerprint quality is determined primarily by area quality and secondary by coordinate axis quality, taking into account the ease of modification and the data capacity required for modification. In other words, if the area quality and the coordinate axis quality are both "excellent", it is determined that no correction is necessary, and the fingerprint quality is defined as "excellent 1".If the area quality is "excellent 6" and the coordinate axis quality is "acceptable" or " If it's not possible.

It was determined that only the coordinate axes needed correction, and the fingerprint quality was
On the other hand, if the area quality is 1 "unsatisfactory", the fingerprint quality is unconditionally defined as "unsatisfactory", and in other cases, the fingerprint quality is defined as "fair".

The fingerprint quality information described above is used to control the flow of fingerprint data. Referring again to FIGS. 1 and 2, the original paper 5 on which the fingerprint data has been read by the fingerprint reader 2 is handed over to the operator. The operator inputs the reading number printed on the back side of the original paper 5 taken into the fingerprint monitor device 3. The image search unit 1d of the central processing unit 1 accesses the magnetic disk device 4 and sends fingerprint data that matches the input reading number to the fingerprint monitor device 3. The operator selects the coordinate axes of the fingerprint data displayed on the fingerprint monitor device 3.

Confirm and correct the extracted contents regarding ZONF, core, and line information. When the confirmation and correction work is completed, the original paper 5 is temporarily stored and handed over to the confirmation process of the collation results.

Next, the flow of fingerprint data according to the fingerprint quality is explained in the 81st iX/
-IFy rlLlQ Valve +101 (;HRRJ-1
The feature point list (including fingerprint quality information) of the fingerprint data of the fingerprint base paper 5 read by the fingerprint reader 2 is stored in the central processing unit 1.
is input into the Using the fingerprint quality as a key from the input minutiae list, the necessary data is input from the fingerprint reading device 2 by referring to the quality table shown in FIG. The fingerprint data input into the central processing bag #1 is stored as is in the original image file 4a. The fingerprint data stored in the original image file 4a will be accessed by the central processing unit 1 from the fingerprint monitor device 3 later.

Monitor display information is displayed on the fingerprint monitor device as shown in FIG. 9 in accordance with the fingerprint quality. If the fingerprint quality is "excellent", the monitor display will display "unnecessary", and there is no need to correct it in this case. If the fingerprint quality is "good", only the coordinate axes need to be corrected, and the feature point list (coordinate axes, ZONE) and skeleton lines are displayed as monitor display information.

If the fingerprint quality is 6”, the coordinate axis.

If it is necessary to modify the ZONE and skeleton lines, the feature point list (coordinate axis, ZONE),
The core line and original image are displayed. Part 1- So, fingers! ? If the ``I quality is spring stiff'', it cannot be corrected, and unnecessary is displayed as monitor display information.

Referring to FIG. 10, in the fingerprint reading device 2.

For a fingerprint whose fingerprint quality is determined to be "excellent", only the feature point list is stored in the feature point file 8 via the original image file 4a and the skeleton file 6. For fingerprints whose fingerprint quality has been determined to be "good," a minutiae list and core lines are displayed on the fingerprint monitor device 3 via the original image file 4a. In the fingerprint monitor device 3.

If the fingerprint data is not modified, only the feature point list is stored in the feature point file 8 via the skeleton file 6. Meanwhile, once the fingerprint data is corrected.

The feature point list and the skeleton are sent to the re-feature extraction device 7 via the skeleton file 6. Then, only the feature point list is output from the re-feature extraction device 7, and this feature point list is stored in the feature point file 8.

For a fingerprint whose fingerprint quality is determined to be 6, the minutiae list, core line, and original image are displayed on the fingerprint monitor device 3 via the original image file 4a.After being displayed on the fingerprint monitor device 3, the original image is The fingerprint data is discarded and the flow of fingerprint data is the same as in the case of "good" fingerprint quality depending on whether or not it is modified.

If the fingerprint quality is "unacceptable", a feature point list is stored in the original image file 4a, and thereafter, the fingerprint data disappears.

If data flow control is performed using fingerprint quality as described above, the capacity of the original image file will be reduced by 70 to 8
0% reduction, and operator correction operation time with the fingerprint monitoring device is reduced by 60-70%.

〔Effect of the invention〕

As explained above, in the present invention, based on the fingerprint quality,
Because it performs fingerprint data flow control.

The capacity of fingerprint data is reduced, which has the effect of shortening the time required for data transfer, and also reduces the disk capacity for temporary storage.

Furthermore, since fingerprint data that does not require correction is not displayed on the fingerprint monitoring device, the operator does not have to judge whether correction is necessary.

There is no need to correct it by mistake. Therefore, the time required for verifying and correcting fingerprint data is shortened, and the quality of the data does not become uneven due to variations in correction due to individual differences.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention, and FIG.
The figure shows the flow of fingerprint base paper, Figure 3 (a) and (b)
are diagrams showing the characteristics of the fingerprints and the matches and discrepancies between the fingerprints, respectively;
FIG. 4(,) is a diagram showing fingerprint data, and FIGS. 4(b) and 4(c) are diagrams for explaining a minutiae list. Figures 4(d) and (,) are diagrams for explaining the core line, Figures 4(f) and (X) are diagrams for explaining the original picture, and Figure 5
Figure (a). (b) and (c) are diagrams for explaining area quality. FIGS. 6(a), (b), and (c) are diagrams for explaining coordinate axis quality, and FIG. 7 is a diagram for explaining fingerprint quality. Figure 8 (,) is a diagram showing the flow of fingerprint data, Figure 8 (b)
) is a diagram showing a quality table, FIG. 9 is a diagram showing the relationship between fingerprint quality, operator interpretation, and monitor display information, and FIG. 10 is a diagram showing the flow n of fingerprint data depending on fingerprint quality. 1... Central processing unit (CPU), 2... Fingerprint reading device. 3...Fingerprint monitor device, 4...Magnetic disk device,
5...Fingerprint paper. Figure 1 Guan Figure 3 ((L) (b>Mim&M(
x, y + tL mountain rrzrrsrra) Engraving of drawing (b) (C) View 8 drawing (b) ○; Large input data procedure supplement 1 book 3 method) j-t3/ε March 2, 1988

Claims (1)

[Claims] 1. Fingerprint reading means for extracting a minutiae list including fingerprint quality information, a core line, and an original image from a fingerprint image; and a fingerprint quality determining means for determining whether to input the core line or the original image based on the fingerprint quality information. , an output means for storing the feature point list in a file means and outputting the core line or original image to the file means according to the determination by the fingerprint quality determining means; an image search means for searching the file means; and an image search means for searching the file means. and a fingerprint monitor means for displaying the search contents retrieved by the means and for confirming and correcting the extracted contents, and controlling the display of the core line or the original image using the fingerprint quality information. Fingerprint data flow control method based on fingerprint quality.