JP5398241B2 - Image processing device - Google Patents

Description

  The present invention relates to a personal authentication technique for performing collation using finger characteristics, and more particularly to a technique for image processing of a fingerprint image for fingerprint authentication.

FIG. 6 shows a configuration of a conventional finger transmission type fingerprint authentication apparatus 600.
A fingerprint photographing unit 601 in the figure photographs a fingerprint.
The image processing unit 602 performs image processing in order to extract information necessary for authentication from the captured fingerprint image.
The storage unit 603 stores registered fingerprint information.
The authentication unit 604 performs fingerprint authentication.
The display unit 605 displays the authentication result.
The traffic information storage unit 606 records the authentication operation.

FIG. 7 shows details of the fingerprint photographing unit 601 in FIG.
As shown in the figure, a fingerprint imaging light source unit 608 installed above the finger emits light, and a camera unit 607 captures the light transmitted through the finger to capture a fingerprint image.

FIG. 8 shows details of the image processing unit 602 of FIG.
A fingerprint extraction unit 6021 in the figure extracts a fingerprint from the input image.
A fingerprint correction unit 6022 performs a correction process for a fingerprint region that has failed to be extracted.
The feature point extraction unit 6023 extracts feature points from the image corrected by the fingerprint correction unit 6022.

Next, the operation will be described.
When the operator captures a fingerprint with the fingerprint capturing unit 601, the captured fingerprint image is sent to the image processing unit 602. When the fingerprint is missing or missing in the minute region in the image processing unit 602, Correct and complete the fingerprint.
Thereafter, feature point information necessary for authentication is extracted, and the information is sent to the fingerprint authentication unit 603.
The fingerprint authentication unit 603 reads fingerprint registration information from the fingerprint registration information storage unit 604 by the number of registration information, and the fingerprint authentication unit 603 compares the fingerprint registration information with the feature point information output from the image processing unit 602 for authentication. Processing is performed, the authentication result is sent to the display unit 606, and the operation record is sent to the travel information storage unit 605.

The conventional finger transmission authentication apparatus has a problem that a vein image is reflected in a fingerprint photographed image as shown in FIG.
In this case, the fingerprint area that overlaps the vein due to the reflection of the vein does not provide sufficient contrast, so accurate fingerprint information cannot be read, and the image may be corrected, resulting in pseudo feature points. There is sex. When such pseudo feature points occur, there is a problem that accurate authentication cannot be performed.
Further, if a human confirms the collected image, it can be determined that the vein is reflected, but this determination cannot be made by image processing. The feature points used for fingerprint authentication are a point where a fingerprint line (ridge line) is branched and a point where a ridge line is broken.

In this regard, Patent Document 1 discloses a technique for performing fingerprint detection processing on the assumption that a fingerprint is always present in an area where a fingerprint is extremely difficult to read.
JP 2007-202912 A

As described above, since a sufficient contrast is not obtained in the fingerprint area overlapping the vein due to the vein appearing in the fingerprint image, accurate fingerprint information cannot be read. However, in Patent Document 1, the fingerprint is very difficult to read. If there is a fingerprint, the fingerprint is corrected and authentication processing is performed assuming that there is always a fingerprint in the area.
There is a high possibility that the fingerprint is restored by correction processing in the region where the fingerprint is missing regardless of the reflection of the vein, but the region where the fingerprint is missing due to the reflection of the vein (black shadow of the vein) is accurate. It is unlikely that it can be restored.
If correction processing is performed on a region where such a vein is reflected, an area with low reliability exists on the fingerprint image.
Then, when using feature points extracted from a fingerprint image in which there are many areas with low reliability, there is a problem that authentication processing cannot be performed with high accuracy.

  One of the main objects of the present invention is to solve the above-described problems, and performs accurate correction processing by determining the presence or absence of the reflection of veins in the captured image, thereby achieving high-precision authentication processing. The main object is to realize an image processing apparatus that can perform the above.

An image processing apparatus according to the present invention includes:
An image input unit for inputting a fingerprint image showing a fingerprint and a vein image showing a vein in the same finger as the fingerprint image;
In the fingerprint image, an area where no fingerprint is reflected is extracted as a fingerprint missing area, and the vein image is collated with the fingerprint missing area, and the fingerprint is complemented with respect to the fingerprint missing area based on the collation result. And a correction processing unit that determines whether or not to perform the correction processing.

  According to the present invention, it is determined whether or not to perform correction processing for complementing the fingerprint on the fingerprint missing area by comparing the vein image against the fingerprint missing area. On the other hand, since there is a high possibility that accurate correction processing can be performed, correction processing is performed, and for fingerprint loss due to reflection of veins, correction processing is not performed because there is low possibility of accurate correction processing. This eliminates correction with low reliability, and enables highly accurate authentication processing.

Embodiment 1 FIG.
FIG. 1 is a block diagram illustrating a configuration example of a finger transmissive fingerprint authentication apparatus 100 according to the present embodiment.
In FIG. 1, a fingerprint / vein photographing unit 1 photographs a vein and a fingerprint.
In the present embodiment, the image processing unit 2 performs image processing on the fingerprint image and vein image captured by the fingerprint / vein imaging unit 1. Details of the image processing unit 2 will be described later.
The other blocks are the same as the functions having the same names shown in FIG.

FIG. 2 shows details of the fingerprint / vein imaging unit 1 of FIG.
As shown in FIG. 2, in this embodiment, the vein imaging light source unit 13 and the fingerprint imaging light source unit 12 installed above the finger alternately emit light, and the camera unit 11 captures the light transmitted through the finger. In this way, a vein image and a fingerprint image are taken.
Thereby, it is possible to obtain a fingerprint image showing a fingerprint and a vein image showing veins in the same finger as the fingerprint image.
For example, the vein imaging light source unit 13 and the fingerprint imaging light source unit 12 are each an LED (Light Emitting Diode).

FIG. 3 shows details of the image processing unit 2 of FIG.
The image processing unit 2 is an example of an image processing device.

The image input unit 20 inputs a fingerprint image captured by the fingerprint / vein imaging unit 1 and a vein image showing the same vein in the finger as the fingerprint image.
The fingerprint / vein determination unit 21 determines whether the input image is a fingerprint image or a vein image.

The correction processing unit 28 extracts an area in which no fingerprint appears in the fingerprint image as a fingerprint missing area, collates the vein image against the fingerprint missing area, and determines the fingerprint for the fingerprint missing area based on the collation result. It is determined whether or not correction processing for complementing is performed.
More specifically, the correction processing unit 28 determines whether or not a vein appears in the position corresponding to the fingerprint missing area in the vein image, and when the vein does not appear in the position corresponding to the fingerprint missing area. The correction process is performed on the fingerprint missing area, and when the vein appears in the position corresponding to the fingerprint missing area, the correction process is not performed on the fingerprint missing area.
In addition, when a vein appears at a position corresponding to the fingerprint missing area, the correction processing unit 28 designates the fingerprint missing area as a feature point extraction non-target area.

The feature point non-target region storage unit 26 stores a feature point non-target region specified by the correction processing unit 28, that is, a region where the vein position and the region where the fingerprint is not shown match.
The feature point extraction unit 27 performs feature point extraction using the fingerprint image corrected by the correction processing unit 28 and the feature point extraction target non-region stored in the feature point extraction target region storage unit 26. .

Next, the internal configuration of the correction processing unit 28 will be described with reference to FIG.
The vein extraction unit 22 extracts a vein from the input vein image.
The vein position specifying unit 23 specifies the position of the extracted vein on the image.
The fingerprint extraction unit 24 extracts a fingerprint from the input fingerprint image.
The fingerprint correction unit 25 collates the fingerprint missing region in which the fingerprint is missing in the fingerprint image (the fingerprint is not shown) and the vein position specified by the vein position specifying unit 23, and the vein appears in the fingerprint missing region. When there is no fingerprint, the correction process is performed on the fingerprint missing area, and when the vein appears in the fingerprint missing area, the correction process is not performed on the fingerprint missing area.

  Next, an operation example of the fingerprint authentication device 100 according to the present embodiment will be described with reference to the flowchart of FIG.

Fingerprint authentication device 100 starts processing upon power-on and waits for a finger to be placed (ST1).
If the finger placement cannot be confirmed (NO in ST2), the process returns to ST1 and the process is repeated.
If the finger placement is confirmed (YES in ST2), the fingerprint / vein imaging unit 1 turns on the vein imaging LED, which is the vein imaging light source unit 13 (ST3), and acquires a vein image (ST4).
Then, in the image processing unit 2, the fingerprint / vein determination unit 21 determines that the acquired image is a vein image, the vein extraction unit 22 extracts a vein image from the vein image, and the vein position specification unit 23 uses the vein The position is specified (ST5), and the vein position is recorded (ST6).

Next, the fingerprint / vein photographing unit 1 turns on the fingerprint photographing LED as the fingerprint photographing light source unit 12 (ST7), and obtains a fingerprint image (ST8).
In the image processing unit 2, the fingerprint / vein determination unit 21 determines that the acquired image is a fingerprint image, the fingerprint extraction unit 24 extracts a fingerprint image from the fingerprint image, and the fingerprint is extracted from the fingerprint image. The missing fingerprint missing area is extracted (ST9).
Next, the fingerprint correction unit 25 reads the position of the vein recorded in ST6, and the vein exists in the corresponding position in the vein image in the fingerprint missing area (the vein in the corresponding position in the vein image). Are extracted) (ST11).
Since this region is a region in which the fingerprint is missing due to the reflection of the vein, the fingerprint correction unit 25 does not perform the fingerprint correction processing that complements the missing of the fingerprint, and the region is excluded from the feature point extraction target region. And stored in the feature point non-extraction region storage unit 26 (ST12).
On the other hand, since the vein does not appear in the region where the vein does not exist in the corresponding position in the vein image (the region where the vein does not appear in the corresponding position in the vein image) in the fingerprint missing region, the fingerprint correction unit 25 Then, fingerprint correction processing for complementing missing fingerprints is performed (ST13).
Next, the feature point extraction unit 27 reads out the feature point non-target region recorded in ST12 (ST14), and extracts feature points for regions other than the feature point non-target region (ST15).
Then, the fingerprint authentication unit 3 reads registration data from the fingerprint registration information storage unit 4 (ST16) and performs fingerprint authentication (ST17).
As a result, when the fingerprint is authenticated (YES in ST18), the lock is unlocked (ST19).
Thereafter, the traffic information storage unit 5 records the operation information (ST20), the display unit 6 displays the authentication result (ST21), and the process ends.

FIG. 5 shows an image diagram of a flow from fingerprint and vein imaging to calculation of a feature point non-target region.
In the example of FIG. 5, there are two fingerprint missing areas (fingerprint missing areas), but the right fingerprint missing area is not corrected because the fingerprint is missing due to the reflection of veins, It is an area not subject to feature point extraction.
By adopting such a configuration, it is possible to suppress the pseudo feature point generated by the overlap between the fingerprint and the vein, which is a problem in the finger transparent authentication, and to improve the authentication performance.

As described above, the fingerprint authentication apparatus described in the present embodiment is a fingerprint authentication apparatus that captures a fingerprint image to be authenticated using transmitted light, and includes a light source for vein imaging and a light source for fingerprint imaging. Using the type, first, a vein is imaged with a vein imaging light source, and then fingerprint imaging is performed with a fingerprint imaging light source.
Next, the vein position on the fingerprint image is calculated in the image processing, and when the fingerprint cannot be clearly photographed on the fingerprint image overlapping with the vein, the correction process is not performed and the area is not used in the authentication process. Also, if the fingerprint cannot be photographed beautifully in the area that does not overlap with the vein image, it is more likely that the fingerprint can be restored by correction processing than in the case where the fingerprint cannot be photographed neatly due to overlapping veins. The correction process is performed as described above and used in the authentication process.
Thereby, it is possible to improve the reliability of the feature point extracted in the black shadow region of the vein reflected when the fingerprint is photographed with the fingerprint photographing light source, and the authentication performance is improved.

Finally, a hardware configuration example of the fingerprint authentication device 100 described in this embodiment will be described with reference to FIG.
As a hardware configuration of the fingerprint authentication device 100, a CPU (Central Processing Unit) 911, a ROM (Read Only Memory) 913, a RAM (Random Access Memory) 914, a magnetic disk device 920, a communication board 915, and the like are connected by a bus 912. .
Further, the CPU 911 is connected to a display device 901, a keyboard 902, a camera 903, an FDD (Flexible Disk Drive) 904, a compact disc device (CDD) 905, a card reader 906, and an electric lock 907.

  In the present embodiment, a program that realizes a function (“˜unit”) shown as an internal element of the fingerprint authentication device 100 is stored as a program group 923 in, for example, the magnetic disk device 920, and the CPU 911 appropriately stores these programs. Each function is executed by reading.

The RAM temporarily stores at least part of a program to be executed by the CPU.
The RAM stores various data necessary for processing by the CPU.

  In the description of the present embodiment, the file group includes “determination of”, “determination of”, “extraction of”, “calculation of”, “comparison of”, “detection of”, “ Information, data, signal values, variable values, and parameters indicating the results of the processing described as “setting of”, “correction of”, “selection of”, etc. are stored in “~ file” or “~ database”. It is stored as each item.

  In addition, the arrows in the flowchart described in the present embodiment mainly indicate input / output of data and signals, and the data and signal values are RAM memory, FDD flexible disk, CDD compact disk, and magnetic disk device. Recording media such as magnetic disks, other optical discs, mini discs, DVDs, and the like. Data and signals are transmitted on-line via buses, signal lines, cables, and other transmission media.

  In addition, what is described as “˜unit” in the description of the present embodiment may be “˜circuit”, “˜device”, “˜device”, and “˜step”, “˜”. “Procedure” and “˜Process” may be used. In other words, what is described as “˜unit” may be realized by firmware stored in the bus. Alternatively, it may be implemented only by software, or only by hardware such as elements, devices, substrates, and wirings, by a combination of software and hardware, or by a combination of firmware. Firmware and software are stored as programs in a recording medium such as a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, and a DVD. The program is read by the CPU and executed by the CPU. In other words, the program causes the computer to function as “to part” of the present embodiment. Alternatively, the procedure or method of “˜unit” in the present embodiment is executed by a computer.

FIG. 3 illustrates a configuration example of a fingerprint authentication apparatus according to Embodiment 1. FIG. 3 is a diagram showing details of a fingerprint photographing unit according to the first embodiment. FIG. 3 is a diagram illustrating details of an image processing unit according to the first embodiment. FIG. 3 is a flowchart showing an operation example of the fingerprint authentication device according to the first embodiment. FIG. 3 is a diagram illustrating an operation principle of the fingerprint authentication device according to the first embodiment. The figure which shows the structural example of the conventional fingerprint authentication apparatus. The figure which shows the detail of the conventional fingerprint imaging | photography part. The figure which shows the detail of the conventional image processing part. The figure explaining the reflection of the vein in a fingerprint picked-up image. FIG. 3 is a diagram illustrating a hardware configuration example of the fingerprint authentication device according to the first embodiment.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Fingerprint and vein imaging | photography part, 2 Image processing part, 3 Fingerprint authentication part, 4 Fingerprint registration information storage part, 5 Passage information storage part, 6 Display part, 11 Camera part, 12 Fingerprint imaging light source part, 13 Light source for vein imaging Unit, 20 image input unit, 21 fingerprint / vein determination unit, 22 vein extraction unit, 23 vein position specifying unit, 24 fingerprint extraction unit, 25 fingerprint correction unit, 26 feature point non-target region storage unit, 27 feature point extraction unit 28 Correction processing unit, 100 Fingerprint authentication device.

Claims (4)

An image input unit for inputting a fingerprint image showing a fingerprint and a vein image showing a vein in the same finger as the fingerprint image;
In the fingerprint image, an area where no fingerprint is shown is extracted as a fingerprint missing area, the vein image is checked against the fingerprint missing area, and a vein appears in the vein image at a position corresponding to the fingerprint missing area. If a vein does not appear in the position corresponding to the fingerprint missing area, a correction process for complementing the fingerprint is performed on the fingerprint missing area, and the position corresponding to the fingerprint missing area is set. An image processing apparatus comprising: a correction processing unit that does not perform correction processing on the fingerprint missing region when a vein is reflected . The correction processing unit
The fingerprint when the vein is reflected in the position corresponding to the missing region, the image processing apparatus according to claim 1, wherein the specifying the fingerprint missing region in the feature point extraction target area outside. The image processing apparatus further includes:
The image processing apparatus according to claim 2 , further comprising a feature point extraction unit that extracts a feature point of a fingerprint for a region other than the feature point extraction target region in the fingerprint image. The image input unit
A fingerprint image captured by using the transmitted light for the fingerprint imaging, according to claim 1 to 3, characterized in that inputs the vein image captured by using the transmitted light for intravenous photographing according to any one Image processing device.

Images