JP2864685B2 - Fingerprint data registration device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Overview] The present invention relates to a fingerprint data registration device, which aims to prevent registration of pseudo feature points due to sweat, dryness, etc., and extracts fingerprint feature points from fingerprint image data. A fingerprint data registration device that cuts out fingerprint image data including feature points in a window shape and registers fingerprint feature point information, the memory storing one or more pseudo feature point image data including pseudo feature points, Window image data cut out in a window shape is subjected to pattern matching with the pseudo feature point image data, and a discrimination unit for discriminating the extracted feature point as a false feature point when the degree of mismatch is equal to or less than a predetermined value, The window image data is subjected to pattern matching with the pseudo feature point image data to determine whether the extracted feature points are true or false. If false, the window image data is excluded from registration. You.

[Industrial applications]

 The present invention relates to an improvement in a fingerprint data registration device.

The fingerprint collation device for identifying an individual is composed of a fingerprint image input device and a collation device, and verifies the identity by comparing fingerprint image data registered in advance with fingerprint image data input from the fingerprint image input device. .

This fingerprint image input device, which is also used at the time of registration, generally uses the optical contact between the finger and the glass, and if the finger is sweating, a bridge due to sweat is created between the fingerprint ridges, as if The fingerprint image is detected such that a branch point exists in the ridge pattern. Conversely, when the finger is dry, the contact between the finger and the glass becomes poor, the ridge pattern is cut, and the fingerprint image is detected as if the end points exist.

Therefore, when such fingerprint image data is registered,
Which can lead to lower matching rates and higher false matching rates.
A fingerprint data registration device that eliminates registration of pseudo fingerprint data due to sweating, drying, or the like is required.

[Conventional technology]

The outline of fingerprint matching will be described with reference to FIG. For details, refer to “Personal verification device using holographic fingerprint sensor” (IEICE PR
U88-38) and the like.

As a method of collating a fingerprint, a branch point of a ridge (a series of convex portions) and a dead end point (referred to as an end point) are set as characteristic points of a fingerprint. At this time, a method of comparing the input fingerprint image data with the personal dictionary information is used.

At the time of registration, first, the ridge portion of the fingerprint image data input from the fingerprint image input device 2 (see FIG. 7) is thinned, and binary data in the vicinity 8 around the point of interest is applied to a feature extraction mask. A plurality of feature points are extracted by the method.

Next, an area including this feature point is cut out to form a window (FIG. 7 to), and the position coordinates of the window, binary data in the window (window image data), and the like are registered as dictionary information. At this time, one window (for example) at the center of the plurality of windows 1 to 3 is defined as a positioning window.

At the time of collation, pattern matching is performed between the window image data and the input fingerprint image data while spirally scanning the alignment window to perform registration between registration and input.
When the positioning is completed, the other windows are translated in accordance with the displacement of the positioning window, and pattern matching is performed in the vicinity of each of the other windows. If the degree of inconsistency of a predetermined number of windows or more is equal to or less than a predetermined value, it is determined that the person is the person.

[Problems to be solved by the invention]

The fingerprint image input device 2 is, as shown in FIG.
A device that detects a fingerprint image by using light reflected from the contact surface 21 by contacting the finger 20 with the finger 22, wherein fingerprint ridges that contact the glass plate 22 and valleys that do not contact are optically separated. The light is received by a sensor (not shown) and binarized.

Therefore, when the finger sweats and a bridge is formed between the fingerprint ridges due to the sweat, as shown in FIG.
The fingerprint image is detected as if a branch point exists in the ridge pattern. Conversely, when the finger is dry, the contact between the finger and the glass becomes poor, and the authenticity of the end point in FIG. As shown in the figure showing an example, it is detected that the ridge pattern is cut as if an end point exists.

When feature points are extracted from fingerprint image data including such pseudo feature points, pseudo feature points may be erroneously extracted, and the matching rate may decrease or the false matching rate may increase.

Therefore, taking advantage of the fact that a large number of true feature points do not exist in the vicinity, for example, when the extracted feature point is an end point, the number of end points in the vicinity is counted, and when the extracted feature point is a predetermined number or more, it is not adopted as a feature point. For example, when the feature point is a branch point, there is a method of determining whether or not the branch point is a pseudo branch point based on whether or not there is a pair of branch points in the vicinity, and removing the pseudo branch point. If the number of points is large, the true end point is excluded from the feature points. In the latter case, if a pseudo branch point exists in the vicinity, the true branch point is determined to be a pseudo branch point and is removed. In any case, there is a problem that true feature points are excluded and the number of feature points required for matching cannot be obtained.

The present invention has been made in view of the above circumstances, and has as its object to provide a fingerprint data registration device for identifying a pseudo feature point.

[Means for solving the problem]

FIG. 1 In the principle diagram of the present invention, reference numeral 9 denotes a memory which stores one or a plurality of pseudo feature point image data 31 including a pseudo feature point.

Reference numeral 4 denotes a discriminating unit, which performs pattern matching on the window image data 30 cut out in a window shape including the extracted feature points with the pseudo feature point image data 31. If the degree of mismatch is equal to or less than a predetermined value, the feature point is determined. Judge as a false feature point.

[Action]

(1) Pseudo feature point image data 31 including pseudo feature points in advance
Is stored in the memory 9. The discriminating unit 4 performs pattern matching of the fingerprint image data cut out in a window shape at the time of registration, that is, the window image data 30 with the pseudo feature point image data 31. If the degree of mismatch is equal to or less than a predetermined value, the feature point is determined. It is determined as a pseudo feature point and is excluded from registration.

(2) As another method of the discrimination, when the matching degree of the pattern matching is equal to or more than a predetermined value, it is discriminated as a pseudo feature point.

(3) When there are a plurality of pseudo feature point image data 31, direction information of ridges or valleys is recorded, and the direction of the ridges or valleys in the window image data is measured, and the corresponding pseudo feature points are measured. Only the image data 31 is subjected to pattern matching.

(4) Similarly, the widths of the ridges or valleys of the pseudo feature point image data 31 are recorded, and the line widths of the ridges or valleys in the window image data are measured. Data 31
Only pattern matching is targeted.

(5) When the area of the pseudo feature point image data 31 is larger than the window area and includes a plurality of pseudo feature points, pattern matching is performed while performing spiral scanning, and the degree of mismatch (match degree) is equal to or less than a predetermined value (or more). Is detected as a false feature point.

The processes (3) to (5) are for improving the processing speed of pattern matching.

As described above, by performing pattern matching with the pseudo feature point image data 31, it is possible to determine whether the feature points are true or false, obtain a required number of feature points, and prevent false registration of false feature points. The drop in the rate and the rise in the rate of incorrect matching are improved.

〔Example〕

 An embodiment of the present invention will be described in detail with reference to the drawings.

2 is a block diagram of one embodiment, FIG. 3 is a flowchart of a registration process, FIG. 4 is a diagram showing a case where a pseudo feature point image data area is larger than a window area, and FIG. 5 is a pseudo feature in a ridge direction. FIG. 6 is an explanatory diagram of selection of point image data. FIG. 6 is an explanatory diagram of selection of pseudo feature point image data based on ridge width.

In the present embodiment, a plurality of pseudo feature point image data 31 are provided,
An example in which the ridge direction and the ridge width are measured to select the pseudo feature point image data 31 will be described.

FIG. 2 shows an example of the configuration of a fingerprint matching device having a registration function. In the figure, reference numeral 1 denotes a numeric keypad for inputting a registration / collation instruction, a registration number, and the like.

A fingerprint image input device 2 inputs fingerprint image data to the collation device 3 in real time.

A collation device 3 has a dictionary information registration function and a fingerprint collation function.

In the collating device 3, reference numeral 8 denotes an image input unit which binarizes the fingerprint image data input from the fingerprint image input device 2 and stores it in the image memory 11.

Reference numeral 6 denotes a registration unit which searches fingerprint image data stored in the image memory 11 to extract a plurality of characteristic points, cuts out a region of a predetermined size including the characteristic points as a window,
The value data (window image data 30) is stored in the window image memory 5, and only the window image data 30 determined as a true feature point is registered in the dictionary memory 10 as dictionary information.

Reference numeral 9 denotes a memory which stores a plurality of pseudo feature point image data 31 in advance.
Are stored together with direction information and line width information.

Reference numeral 4 denotes a discrimination unit that performs pattern matching between the clipped window image data 30 and pseudo feature point image data 31 and, if the degree of mismatch is equal to or less than a predetermined value, discriminates the pseudo feature point and notifies the registration unit 6.

Reference numeral 7 denotes a collating unit which reads corresponding dictionary information from the dictionary memory 10 based on the registration number input from the ten key 1,
Check with the input fingerprint image data.

 With the above configuration, the following registration processing is performed.

(1) Set the registration mode using the numeric keypad 1, enter the registration number, and touch the finger 20 on the contact surface of the fingerprint image input device 2.
Thus, the fingerprint image is read, converted into binary data, and stored in the image memory 11.

(2) The registration unit 6 searches the fingerprint image data stored in the image memory 11 and extracts feature points. Then, data of a predetermined area including the extracted feature points is cut out and stored in the window image memory 5 as window image data 30.

(3) The determination unit 4 performs a feature point determination process described below.
The authenticity is notified to the registration unit 6.

First, a thinning process is performed to erase one dot at a time around the ridge in the window image data 30, and the line width is measured by a method such as counting the number of thin lines, and based on the final thinned image data. Measure the direction.

(4) The corresponding pseudo feature point image data 31 is selected based on the measured direction and line width.

FIG. 5 shows a case in which pseudo feature point image data classified into two directions is selected. The ridge direction in the window image data 30 is recognized by measuring the gradient of the thinned image data or the like. This indicates that pattern matching is performed by selecting pseudo feature point image data.

Since the fingerprint is spiral, the ridge direction can be classified into multiple directions, but if it is classified into multiple directions such as horizontal and vertical,
The pseudo feature point image data 31 close to the measurement result is selected.

FIG. 6 shows an example of selecting pseudo feature point image data having different line widths. The line width is measured in the process of thinning, and pseudo feature point image data having a similar line width is selected.

In the present embodiment, since both the direction and the line width are measured and selected, if the pseudo feature point image data 31 is set to, only the pseudo feature point image data 31 is selected, and the processing time of pattern matching, that is, the determination time is greatly reduced. Be shortened.

(5) Perform pattern matching for detecting matching / mismatch of corresponding pixels between the window image data 30 and the selected pseudo feature point image data 31, count the number of mismatches, and obtain the degree of mismatch.

(6) If the degree of inconsistency is equal to or smaller than the threshold value, it is determined that the pseudo feature point image data 31 matches, and the registration unit 6 is notified.

(7) The registration unit 6 registers the window image data 30 determined as a true feature point in the dictionary memory 10 together with the position information as dictionary information, and extracts the next feature point.

Thereafter, the same determination and registration processing is performed, and after the number of true feature points required for collation are extracted, the registration processing ends.

FIG. 4 shows an example in which the pseudo feature point image data area is made larger than the window area and a plurality of pseudo feature points are included, which has the effect of reducing the number of pseudo feature point image data. in this case,
As shown in the figure, the window image data 30 is converted to pseudo feature point image data.
Pattern matching is performed while scanning spirally from the approximate center of 31, the degree of mismatch at each position is calculated, and it is determined by detecting whether or not there is a threshold value or less.

Although all of the above were determined by calculating the degree of mismatch,
The same result can be obtained by discrimination based on the degree of coincidence.

As described above, the authenticity of the extracted feature points can be determined by performing pattern matching with the pseudo feature point image data 31, and the number of feature points required for collation can be determined from the collected fingerprint image data regardless of the wetness or dryness of the finger. It can be extracted, and the false feature point is not erroneously registered to lower the matching rate, and the false matching rate is not increased.

〔The invention's effect〕

As described above, according to the present invention, the authenticity of the extracted feature point is determined and registered by pattern matching with the pseudo feature point image data, and the true feature point is extracted without being affected by the dryness and humidity of the finger. Since registration can be performed, it is possible to prevent a reduction in the matching rate and an increase in the erroneous matching rate, and it is possible to extract a required number of feature points regardless of the wetness or dryness of the finger.

[Brief description of the drawings]

FIG. 1 is a principle diagram of the present invention, FIG. 2 is a configuration diagram of one embodiment,
FIG. 3 is a flowchart of the registration process, FIG. 4 is a diagram showing a case where the pseudo feature point image data area is larger than the window area, FIG. 5 is an explanatory diagram of selection of the pseudo feature point image data in the ridge direction, and FIG. Is an explanatory diagram of selection of pseudo feature point image data based on a ridge width, FIG. 7 is an explanatory diagram of a fingerprint registration / collation method, FIG. 8 is a diagram showing an example of authenticity of an end point, and FIG. FIG. In the figure, 1 is a numeric keypad, 2 is a fingerprint image input device, 3 is a collation device, 4 is a discriminating unit, 5 is a window image memory, 6 is a registration unit, 7 is a collation unit, 8 is an image input unit, 9 is a memory, Reference numeral 10 denotes a dictionary memory, 11 denotes an image memory, 20 denotes a finger, 21 denotes a contact surface, 22 denotes a glass plate, 30 denotes window image data, and 31 denotes pseudo feature point image data.

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G06T 7/00

Claims (5)

(57) [Claims] 1. A fingerprint data registration apparatus for extracting a feature point of a fingerprint from fingerprint image data, extracting the fingerprint image data including the feature point in a window shape, and registering fingerprint feature point information, comprising: A memory (9) storing one or more pseudo feature point image data (31) including the window image data (30) cut out in a window shape and pattern matching with the pseudo feature point image data (31). And a discriminator (4) for discriminating the extracted feature point as a false feature point when the degree of mismatch is equal to or less than a predetermined value, and performing pattern matching on the window image data with the pseudo feature point image data. A fingerprint data registration apparatus for determining the authenticity of the feature point extracted in step (a), and excluding the window image data from the registration if the feature point is false. 2. The fingerprint data according to claim 1, wherein said pattern data is a discriminator for discriminating the extracted feature point as a false feature point when the degree of coincidence is equal to or greater than a predetermined value. Registration device. 3. A method for detecting a direction of a fingerprint ridge or a valley in the window image data, selecting pseudo feature point image data corresponding to the detected direction from the plurality of pseudo feature point image data, and performing pattern detection. The fingerprint data registration device according to claim 1, wherein the fingerprint data registration device is a determination unit that performs matching. 4. A method for measuring a line width of a fingerprint ridge line or a valley line in the window image data, selecting pseudo feature point image data corresponding to the line width from the plurality of pseudo feature point image data, and selecting a pattern. 4. The fingerprint data registration device according to claim 1, wherein the fingerprint data registration device is a discriminating unit that performs matching. 5. When the pseudo feature point image data has an area larger than the window area and includes a plurality of pseudo feature points, pattern matching is performed while scanning the window image data within the area, and the degree of mismatch is determined. 5. The fingerprint according to claim 1, wherein the discriminating unit discriminates a pseudo feature point when the degree of coincidence is equal to or less than a predetermined value. Data registration device.