JPH077447B2 - Fingerprint matching device - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a fingerprint collation device for directly collecting and collating an uneven pattern of a fingerprint in real time.

[Conventional technology]

FIG. 5 is a system diagram showing a conventional fingerprint collation device disclosed in, for example, Japanese Patent Laid-Open No. 61-145686.
Denotes a light source, 2 denotes an optical device such as a right-angle prism, and an incident light flux L ₁ from the light source 1 is incident on the slope 2a on which the finger F of the subject is placed from the surface 2b at a predetermined angle within a certain range.

Reference numeral 3 denotes a conversion device such as a television camera, which converts the optical information (reflected light beam bundle L ₂ ) output from the slope 2a of the optical device 2 into an electric signal to obtain a fingerprint image.

Reference numeral 4 denotes an automatic discriminating device, which automatically discriminates whether or not the fingerprint image can be taken in from the converting device 3 based on the average density.

5 notifies the subject of the discrimination result of the automatic discrimination device 4,
For example, a display device having three LEDs 5a, 5b, 5c is shown, and the automatic display device 4 and the display device 5 constitute a display device.

Reference numeral 6 denotes a processing device for identifying and collating the captured fingerprint image.

Next, the operation will be described.

First, in an initial state in which the finger F is not placed on the slope 2a of the optical device 2, the LED 5a of the display device 5 is turned on and gives an instruction to the subject to place the finger F.

Next, when the subject puts his / her finger F on the slope 2a, the LED 5a turns off and L
The ED5b flashes and the incident light flux L ₁ from the light source 1 is transmitted to the optical device 2
Since the light is incident on the inclined surface 2a from the surface 2b at a predetermined angle exceeding the critical angle θ, total reflection is performed at the interface between the inclined surface 2a of the optical device 2 and air, and at the interface where the finger F is in close contact with the medium. Due to the refractive index relationship of, total reflection is stopped. That is,
The portion of the fingerprint valley line Fa that does not adhere to the slope 2a is totally reflected, and the portion of the fingerprint ridge Fb that adheres to the slope 2a is not totally reflected.

Therefore, the reflected light bundle L ₂ output from the surface 2c of the optical device 2 is converted into a fingerprint image by the conversion device 3, and the automatic determination device 4 determines whether the contrast of the fingerprint image from the conversion device 3 is clear. To do. At this time, if it is determined that the contrast is not clear, the process waits until the surface of the finger F gets wet.

Further, when the automatic discriminating device 4 discriminates that the contact between the finger F and the slope 2a is increased and it is a concavo-convex pattern of a clear fingerprint image with contrast, the fingerprint image is taken into the processing device 6 at that time and the LED 5b Disappears and the LED 5c lights up to inform the subject that a clear fingerprint image has been captured by the processing device 6.

Then, the processing device 6 performs processing such as identification and matching of the captured fingerprint image, so that the subject can release the finger F from the slope 2a.

When these processes are completed, the LED 5c disappears, the LED 5a lights up again, and the initial state is restored.

Next, with reference to FIG. 6, a description will be given of the determination of presence / absence of contrast performed by the automatic determination device 4 described above.

First, as a pre-step of fingerprint collation, the image P ₁ in a state where the finger F is not placed (FIG. 6 (a)) is captured and four regions S ₁
˜S ₄ Concentration variance σ _i ⁽¹⁾ (i = 1 to 4) is calculated.

Next, the image P ₂ of the state in which the finger F is placed (FIG. 6 (b)) is captured, and the variance σ _i of the density in the _four regions S _{1 to} S ₄ is taken.
⁽²⁾ Calculate (i = 1 to 4).

And R: A finger F is placed, or an index for determining whether there is contrast is calculated.

Next, if R ≦ T ₁ (T ₁ : threshold value for determining whether the finger F is placed) is satisfied, it is determined that the finger F is not placed, and T ₁
If <R ≦ T ₂ (T ₂ : threshold for judging whether the contrast is appropriate) is satisfied, it is determined that the surface of the finger F is dry, and if T ₂ ≦ R is satisfied, the contrast is high. Judge that an image was obtained.

[Problems to be solved by the invention]

Since the conventional fingerprint collation device is configured as described above, that is, by comparing the image P ₂ with the finger F placed with the image P ₁ without the finger F, the fingerprint image Since it is configured to determine the presence / absence of contrast, it becomes difficult to set the capture timing of the image P _{1 with} the finger F not placed.

Therefore, for example, if a fingerprint image in which the surface of the slope 2a is soiled with oil is captured, the appropriate density distribution σ _i
There is a problem that the value of ⁽¹⁾ cannot be obtained, and the index R cannot be a correct index for determining the presence or absence of contrast.

Further, if the subject is configured to judge the timing of capturing the fingerprint image, there is a problem that the entire operation becomes complicated.

The present invention has been made to solve the above-mentioned problems, and determines whether or not the subject's finger is placed, and whether or not there is contrast in the fingerprint image is determined from the shape of the density histogram of the image. An object of the present invention is to obtain a fingerprint collation device that can collate fingerprints accurately.

[Means for solving problems]

The fingerprint collation device according to the present invention is configured to determine the presence or absence of contrast in a fingerprint image based on the shape of the density histogram.

[Action]

The fingerprint collation device according to the present invention discriminates the presence or absence of the contrast of the fingerprint image by the shape of the density histogram and fetches it.

〔Example〕

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

In FIG. 1, the same parts as those in FIG. 5 are designated by the same reference numerals, and 7 indicates an automatic discriminating device, which automatically discriminates the suitability of the fingerprint image from the converting device 3 in the shape of a density histogram. Is.

8 informs the subject of the discrimination result of the automatic discrimination device 1,
For example, a display device having three LEDs 8a, 8b, 8c is shown, and the discrimination display device is constituted by the automatic discrimination device 7 and the display device 8.

FIG. 2 is a block diagram showing the configuration of the automatic discrimination device,
Reference numeral 71 is an image memory for storing image data (fingerprint image) from the conversion device 3, 72 is a histogram calculation circuit, and 73 is an index R.
R is a calculation circuit (distance degree calculation circuit), 74 is a ω ₁ calculation circuit (ratio calculation circuit) that calculates the ratio ω ₁ of “black” pixels, and 75 is a threshold value T ₁ , T ₂₁ and T ₂₂ . The stored ROMs, 76a, 76b, and 76c are comparators (first and second comparators) 77a,
Reference numeral 77b is an AND circuit, 78 is a NOT circuit, and 79a and 79b are NAND circuits.

In addition, "L" is predominant in each signal, and "H" is insignificant.

Next, the flowchart of FIG. 3 and FIGS. 4 (a) to 4 (d).
The operation will be described with reference to the characteristic diagram of the density histogram of FIG.

First, when the signal ▲ ▼ is set to “L”, the signal ▲
▼ is "L", ▲ ▼ and ▲
The ▼ becomes “H” and turns on the LED 8a of the display device 8 and turns off the LEDs 8b and 8c to prompt the subject to place the finger F on the slope 2a of the optical device 2 (step ST
(1)), the image of the slope 2 is loaded into the image memory 71 via the conversion device 3 (step ST (2)), and the histogram calculation circuit 72 creates a density histogram in a predetermined small area (step ST ( 3)) and the shape of the density histogram is as shown in FIGS. 4A to 4D depending on whether the finger F is placed and whether there is contrast. For example, FIG. 4 (a) shows a state in which the finger F is not placed, FIGS. 4 (b) to 4 (d) show a state in which the finger is placed, and FIG. 4 (b) shows "black", When the "white" pixels are appropriate and have a good contrast, FIG. 4C shows that when the finger F is too moist or the finger F is held too strongly, the ratio of "black" pixels is large. (D) is the reverse case of FIG. 4 (c).

Then, for ω ₁ and ω ₂ , the density is divided into two classes, “black” and “white”, so that each class has the best cohesion.
The ratio of “black” pixels and “white” pixels when the threshold value t is set is shown, and (ω ₁ + ω ₂ ) = 1.

Further, μ ₁ and μ ₂ represent average densities of “black” and “white” classes.

After all, a clear image with good contrast means an average density μ ₁ ,
It can be considered that μ ₂ is separated and the ratios ω ₁ and ω ₂ are approximately the same.

Next, the R calculation circuit 73 calculates R = (μ ₂ −μ ₁ ) / μ ₂ (step ST (4)), and the comparator 76a compares the index R with the threshold value T _1, and R <T _{If 1} , that is, if the output of the comparator 76a is "L" (as shown in FIG. 4 (a)), it is determined that the finger F is not placed (step ST
(5)), return to step ST (1), and if R ≧ T ₁ (in the case of FIG. 4 (b) to (d)), it is determined that the finger F is placed and the LED 8b is turned on. At the same time, the LED 8a is turned off (step ST (6)).

When the R calculation circuit 73a and the ω ₁ calculation circuit 74 operate once, the signal ▲ ▼ becomes "H".

Next, the comparator 76b calculates the ratio ω ₁ and the threshold value T ₂₁ by
The ratio ω ₁ is compared with the threshold value T ₂₂ by the comparator 76c to determine whether T ₂₁ > ω ₁ and T ₂₂ <ω ₁ (T ₂₁ , T ₂₂ : contrast is proper together with step ST (5). Threshold value), that is, if the outputs of the comparators 76b and 76c are both "L" (4th
In the case of (c) and (d), it is determined that the contrast is not appropriate (step ST (7)), and step ST
Returns to (2), (case shown in FIG. 4 (b)) If _{T 21 ≦ ω 1 ≦ T 22} , with turns off the LED8b to determine the contrast is in a suitable range, to light the LED8c (Step ST (8)), the image is captured and sent to the processing device 6, and at the same time, the processing is notified to the processing device 6 by the signal ▲ ▼.

It should be noted that although the creation of the density histogram is described as one small area in the above embodiment, the judgment is actually made from the shapes of the density histograms of a plurality of small areas.

Also, if the ratio ω ₁ of “black” pixels is displayed in a bar graph or the like, the subject can know the moistness of the finger and the strength of the holding method. Can be adjusted to be appropriate.

〔The invention's effect〕

As described above, according to the present invention, since the adequacy of the contrast of the fingerprint image is determined by the density histogram, a clear image with contrast can be captured,
It has an excellent effect that accurate fingerprint matching can be performed.

[Brief description of drawings]

FIG. 1 is a system diagram showing a fingerprint collation device according to an embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of an automatic discrimination device, FIG. 3 is a flow chart showing the operation, and FIG. 4 (a).
(D) is a characteristic diagram of a density histogram, FIG. 5 is a system diagram showing a conventional fingerprint collation device, and FIG. 6 is an explanatory diagram for judging contrast. In the figure, 1 is a light source, 2 is an optical device, 3 is a conversion device,
6 is a processing device, 7 is an automatic discriminating device, and 8 is a display device. In the drawings, the same reference numerals indicate the same or corresponding parts.

Claims (2)

[Claims] 1. An optical device in which a light beam from a light source is incident on a surface on which a finger of a subject is placed at a predetermined angle, and optical information from the optical device is converted into an electric signal to form an uneven pattern of a fingerprint. To obtain a histogram of the fingerprint image obtained by this conversion device, the average density of the black class and the average density of the white class of the histogram is more than a predetermined value, and the black pixel and the white pixel When the ratios are the same, a discrimination display device that discriminates that there is contrast and notifies the subject, and a processing device that captures the fingerprint image that is discriminated by the discrimination display device and performs collation. A fingerprint matching device provided. 2. The discrimination display device includes an image memory for storing a fingerprint image from the conversion device, a histogram calculation circuit for obtaining a histogram from the fingerprint image, and an average density of black class and an average density of white class from the histogram. Distance calculation circuit for calculating an index indicating the separation degree, a first comparator for comparing the calculated index indicating the separation degree with a reference value of the separation degree, and a black pixel and a white pixel from the histogram. The fingerprint collation device according to claim 1, further comprising: a ratio calculation circuit that calculates a ratio of the ratio and a second comparator that compares the calculated ratio with a reference value of the ratio. .