KR102015118B1 - A Vessels Pattern Recognition Based Biometrics Machine using Laser Speckle Imaging and Methods Thereof - Google Patents

Abstract

The present invention relates to a vein pattern analysis of the biometric authentication target site using a laser speckle imaging method and a biometric authentication device through blood flow information, and more specifically, a sensor for recognizing the presence of the biometric authentication target site in a receiving space; A laser irradiation unit for irradiating a laser onto the biometric authentication target site; An image generator for converting light reflected by the laser irradiation into an image through a laser speckle imaging method; An imaging sensor to capture the converted image; A biometric determination unit that determines whether the authentication target part is a living body by verifying whether the information value input from the imaging sensor is within a biometric information value range; An image processor extracting vein pattern information from the captured image through a noise filtering algorithm and displaying the vein pattern as an image having improved accuracy; And a biometric authentication unit configured to perform user authentication by comparing the image having the improved accuracy and the vein pattern image previously stored in the memory.

Description

A Vessels Pattern Recognition Based Biometrics Machine using Laser Speckle Imaging and Methods Thereof}

The present invention relates to a biometric authentication device using a vein pattern analysis using a laser speckle imaging method and a biometric authentication method using the same, and more particularly, to improve accuracy and security by applying a speckle imaging technique. The present invention relates to a biovascular authentication-based biometric authentication device through biometric determination and vein pattern information recognition and a biometric authentication method using the same.

Biometrics is defined as the identification of identity by automatically measuring physical or behavioral characteristics. Domestic biometric forums define human perception based on observation of behavioral and biological (anatomical and physiological) features.

Body information is extracted directly from the human body. Fingerprint, Iris-scan, Retina-scan, Hand geometry, Facial recognition There is a way. Methods using behavioral characteristics include voice recognition, signature-scan, and the like.

The use of voice, retina, iris, face, and fingerprints in biometric systems has already been developed, and even in the UK, methods of using human odors have been developed. Biometrics uses people's unchanged physical or behavioral characteristics, such as fingerprints and voices, to identify them. As far back as ancient Egypt, the pharaohs measured the heights of men to identify individuals. Fingerprinting has been in use for more than a century. Biometric equipment in the late 1960s and early 1970s used human authentication to measure the length of a finger.

The global market for biometric devices and systems was close to $ 100 million by 1999, with 76 percent of the total demand supplied to the United States. Using biometrics has many advantages. It is becoming more and more frequently used. The apartments in Chicago use a fingerprint recognition system for entrance and exit, Disneyland abolished the ID card that used existing photographs, replaced it with fingerprint recognition, and financial institutions such as the US bank, Citybank, also recognized the iris. You are trying to adopt a system.

Recently, an authentication method using multiple elements of a living body that combines other biometric authentication based on authentication using a vein pattern has been proposed. Two-factor authentication combining a palm vein pattern and a palm contour shape (Korean Patent Publication No. 10-2010-152706), and a technology for authenticating using a palm palm shape and a finger vein shape have been proposed (Korean Patent No. 1,517,371). . However, because the palm of the finger or palm is the information on the surface of the human body, it is not only easy to acquire the characteristics, but also has the disadvantage that the personal authentication by the palm is vulnerable to forgery, thereby improving the accuracy of biometric authentication.

Therefore, the present inventors have made efforts to improve the accuracy and security of the vein pattern image, and as a result, the biometric authentication of the individual through the vein pattern of the biometric authentication target region represented by the image with improved accuracy using the laser speckle imaging method It has been confirmed that the present invention has been completed.

An object of the present invention is to provide a bio-vascular recognition-based biometric authentication device including a sensor, a laser irradiation unit, an image generating unit, an imaging sensor, a biometric determination unit, an image processing unit, a biometric authentication unit.

Another object of the present invention is to provide a biometric authentication method using a biometric device.

In order to achieve the above object, the present invention (a) a sensor for recognizing the presence of a biometric authentication target site in the accommodation space and the receiving space for receiving the biometric authentication target site in a non-contact manner; (b) a laser irradiation unit for irradiating a laser to the biometric authentication target site; (c) an image generator for converting light reflected by the laser irradiation into an image through a laser speckle imaging method; (d) an imaging sensor to capture the converted image; (e) a biometric determination unit configured to determine whether the information value input from the imaging sensor is within a pre-registered biometric value range and determine whether the authentication target site is a living body; (f) an image processor extracting vein pattern information from the captured image through a noise filtering algorithm and displaying the vein pattern as an image having improved accuracy; And (g) provides a bio-vascular recognition-based biometric authentication device comprising a bio-authentication unit for performing a user authentication by comparing the image of the improved biometric determination and accuracy and the pre-stored vein pattern image in the memory.

The present invention also includes the steps of: (a) converting the light reflected by the laser irradiation to the biometrics target site into an image through a laser speckle imaging method; (b) capturing the converted image; (c) determining whether an authentication target site is biometric based on the input information value; (d) extracting vein pattern information from the captured image through a noise filtering algorithm to represent the vein pattern with improved accuracy; And (e) comparing the vein pattern image stored in the memory with the image of which the biometric determination and accuracy are improved and performing user authentication.

According to the present invention, since blood flow information based on laser speckle can be additionally registered to determine whether the living body is present, authentication attempts using the replication of pseudo vein patterns can be eliminated at the source. It is excellent and does not have to worry about loss like ID card and can be recognized without touching the recognition device and palm, so it is safe from bacterial infection in public places, and can be used even in vulnerable condition due to sweat or oily skin.

Figure 1 shows a schematic diagram of a biometric system capable of recognizing vein pattern and vein blood flow information of the palm by irradiating the laser light scattered uniformly from above.
FIG. 2 is a schematic diagram of a biometric system capable of recognizing both vein patterns and vein blood flow information of the back of the hand and the palm by uniformly scattering laser light in both directions below and above.

The present invention is applied to the laser speckle imaging method by applying a filtering algorithm that can extract the pattern information of veins having a thickness of 1 mm or more and 2 mm or more based on the average 1.27 mm hand and the dorsal venous network of hands. This improved vein image is obtained using a near-infrared light source of 700 nm to 1000 nm, which is relatively absorptive in the epidermis and dermis of the skin, and at the same time, the biometric authentication technology enhances security through biometric determination using blood flow information. It is about.

The present invention, in one aspect, (a) a sensor for recognizing the presence of the biometric authentication target site in the accommodation space and the receiving space for receiving the biometric authentication target site in a non-contact manner; (b) a laser irradiation unit for irradiating a laser to the biometric authentication target site; (c) an image generator for converting light reflected by the laser irradiation into an image through a laser speckle imaging method; (d) an imaging sensor to capture the converted image; (e) a biometric determination unit configured to determine whether the information value input from the imaging sensor is within a pre-registered biometric value range and determine whether the authentication target site is a living body; (f) an image processor extracting vein pattern information from the captured image through a filtering algorithm and displaying the vein pattern as an image having improved accuracy; And (g) relates to a bio-vascular recognition-based biometric authentication device comprising a biometric authentication unit for performing the user authentication by comparing the image with the improved accuracy and the pre-stored vein pattern image in the memory.

Hereinafter, embodiments of a biometric authentication device and a biometric authentication method according to the present invention will be described in detail with reference to the accompanying drawings.

1 and / or 2, the biometric authentication target site accommodating space, the sensor, the laser irradiation unit, the image generating unit, the imaging sensor, the biometric determination unit, the image processing unit, and the biometric authentication unit are configured. The laser irradiation unit may be positioned above and / or below the biometric authentication target site.

The biometric subject site preferably means the back of the hand and / or the palm.

The biometric subject area receiving space is for accommodating in a non-contact manner and may be implemented as an insertion hole into which a hand is inserted. In addition, a sensor may be provided at the entrance of the accommodation space to detect the presence of a hand to activate the operation of the laser irradiator, the image sensor, the biometric determination unit, the image generator, the image processor, the memory, and the biometric authentication unit.

The laser irradiation part is for irradiating a linear laser light extending in the longitudinal direction of the back of the hand and / or palm to the back of the hand and / or the palm accommodated in the back of the hand and / or the palm accommodation space, and may include a near infrared ray or an LED light source. Preferably, the light source is a near infrared light source having a wavelength of 700 nm to 1000 nm having relatively low absorption in the epidermal and dermal areas of the skin, and more preferably a near infrared light source having a wavelength of 740 nm and 850 nm.

In addition, it is preferable that the laser light source including a diode laser, VCSEL, etc. having high coherence of the light source for speckle formation.

In the related art, a hygiene problem may occur because a finger needs to be in close contact with a light source due to a problem in performance. As it is excellent, this conventional problem does not occur.

The image generator is to convert the light irradiated from the laser irradiator into an image through a laser speckle imaging method. A laser speckle is a grain-shaped irregular pattern created by coherent light or light reflected when a laser is injected into a scattering material. Laser speckle imaging technology allows imaging of these phenomena and image processing using statistical properties to allow for the observation of blood vessels.

In order to observe the blood vessels of the laser speckle imaging, an image image processing method is required, LASCA (Laser Speckle Contrast Analysis), mLSI (Modified Laser Speckle Imaging), tLASCA (Temporally LASCA), sLASCA (Spatially LASCA), anisotropic processing, and It may be selected from the group consisting of a combination of these, and preferably a combination of a temporal technique and a spatial filter technique.

The imaging sensor includes a camera that captures an image generated by the image generator. Since the image sensor photographs incident light, the image sensor may be implemented using a small near infrared filter, and may be compactly implemented as a CIS image sensor using a semiconductor process.

The camera is for photographing a hand accommodated in the back of the hand and / or the palm accommodation space, and is preferably a camera (CCD camera) capable of converting an image into an electrical signal. The back of the hand and / or palm can be installed in a suitable position to be clear, it is preferable to be installed in a position forming a 90 ° ~ 180 ° with the laser irradiation around the back of the hand and / or palm receiving space. For example, it may be provided opposite or to the side of the laser irradiation part. This installation location may be appropriately changed depending on the overall structure of the biometric authentication system.

The principle of the biometric authentication system using the vascular pattern is that the absorbance of the blood in the near infrared wavelength region is higher than that of the surrounding biological tissues, and when the light in the near infrared wavelength region is reflected on the back of the hand and / or the palm, the portion of the blood vessel passing by the blood is dark. It is shaded to show blood vessel patterns on the back of the hand and / or palm. When the light source emits wavelengths in the near infrared region, the blood vessel patterns of the back of the hand and / or palm can be observed in more detail.

Therefore, the camera is preferably a camera having a high sensitivity in the visible light region and the near infrared region that can detect the wavelength of the near infrared band where blood red blood cells mainly absorb light. Also, in order to obtain a clear blood vessel pattern image, a near infrared filter (IR filter) mainly passing light in the near infrared band may be installed in front of the camera.

The biometric determination unit may be obtained through image processing of reflected light or transmitted light of a body part attempting authentication by a highly coherent light source on an image collecting device, a CCD, or a CMOS. Preferably, the image processing means blood flow information. The method may include verifying whether the blood flow information is within a specific numerical range registered in advance.

The image processor extracts a blood vessel pattern by analyzing and processing an image input from a camera and stores the extracted blood vessel pattern, and may be a tablet pc, an embedded system, a PDA, a mobile phone, a computer, a microprocessor, a microcomputer, or a microcontroller. And other associative devices.

The image processing unit may extract vein pattern information by a filtering algorithm, and may extract vein pattern information based on 1 to 2 mm, which is a general vein thickness range. The average hand vein thickness is 1.27 mm on average, and even thinner veins of less than 1 mm are available. However, if a filtering algorithm capable of extracting veins with a thickness of 1 mm or more is used, a more efficient vein recognition system can be constructed. have.

The image processor may control operations of the laser irradiator and the camera based on an image input from the camera. For example, when the blood vessel pattern of the back of the hand and / or the palm photographed by the camera is impossible to read due to lack of light or excessive light, a control signal is applied to adjust the light amount of the light source.

In addition, the image processor may adjust the camera exposure time in addition to adjusting the amount of light of the light source. The image processing unit increases the exposure time of the camera together with adjusting the amount of light of the light source when the blood vessel pattern of the back of the hand and / or palm of the hand is insufficient in light, and reduces the exposure time of the camera in the case of excessive light, and / or the hand for easy reading Control to acquire the vascular pattern of the palm.

For example, the image processor may easily obtain a clear and identifiable image regardless of the difference in the thickness of the human hand by controlling the light amount of the light source and the exposure time of the camera according to the brightness and contrast of the captured image. For example, when the back of the hand and / or palm is thick, the output of the light source is increased or the exposure time of the camera is increased, and when the back of the hand and / or palm is thin, the back of the hand and / or palm is reduced by lowering the output of the light source or reducing the exposure time of the camera. The effect of the thickness difference can be reduced.

The memory registers and stores the vein pattern image of the authenticated person.

The biometric authentication unit compares the image pattern acquired through the image generation unit with the vein pattern image registered in the memory and determines whether to authenticate according to a matching result. For example, the image generating unit extracts and stores a blood vessel pattern from an image image photographed during a user registration process, and then processes a user authentication by comparing blood vessel patterns extracted from the captured image with blood vessel patterns previously stored in a memory.

The biometric authentication of the present invention is characterized in that the user is authenticated only when the blood flow information is included in the biometric range and at the same time the vein pattern is matched.

1 and 2 of the present invention to describe the operation of the biometric authentication device by recognizing the back of the hand and / or palm vein pattern.

First, the user inserts the back of the hand and / or the palm of the biometric authentication target portion into the biometric authentication device. The presence of the hand in the receiving space is transmitted to the computer by the sensor. Then, the authentication target site is irradiated with laser light uniformly scattered below and / or above, and the light reflected thereby is converted into a blood flow image through a laser speckle imaging method and captured. The biometrics of the authentication target site is determined based on the input blood flow information. User authentication is performed by comparing the extracted image with the vein pattern image previously stored in the memory. Blood flow information obtained through biometric determination and speckle imaging is extracted to determine final authentication based on additional vein pattern images and mutual matching results.

Accordingly, the present invention in another aspect, (a) recognizing the presence of the biometric authentication target site through a sensor in the receiving space for accommodating the biometric authentication site in a non-contact manner; (b) irradiating a laser to the site of biometric authentication; (c) converting the light reflected by the laser irradiation into an image through a laser speckle imaging method; (d) capturing the converted image; (e) extracting vein pattern information from the captured image through a noise filtering algorithm to represent the vein pattern with improved accuracy; And (f) performing user authentication by comparing the vein pattern image pre-stored in the memory with the image of which the biometric determination and accuracy are improved.

Hereinafter, the present invention will be described in more detail with reference to Examples. These examples are only for illustrating the present invention, it will be apparent to those skilled in the art that the scope of the present invention is not to be construed as being limited by these examples. Thus, the substantial scope of the present invention will be defined by the appended claims and their equivalents.

Example 1: Biometric Identification System Based on Laser Speckle Imaging

1 is a schematic diagram of a biometric system capable of recognizing a vein pattern of a palm by irradiating laser light scattered uniformly from above. FIG. 2 is a schematic diagram of a biometric system capable of recognizing vein patterns of the back of the hand and the palm by irradiating laser light uniformly scattered in both the lower and upper directions. With reference to this will be described biometrics-based biometric authentication method.

Biological vascular recognition based biometric authentication system includes a sensor capable of identifying the presence of a hand, a laser light source that can be irradiated below and / or above, a camera capable of capturing vein images of the back of the hand and / or palm, a biometric unit, memory, and a user authentication unit. It is configured by. A sensor capable of confirming the presence of a hand may be detected through movement of an entrance surface, as shown in FIGS. 1 and 2, rather than in contact with a slide. The laser light that can be irradiated below and / or above is used as a light source, and uniformly scattered light can be irradiated to show a vein pattern. In this case, the light source is located in the opposite direction of the image sensor, and the vein pattern and speckle through the transmitted and scattered light are displayed. You can get a video.

In addition, the vein image of the back of the hand and / or palm was captured and imaged using a CCD camera capable of converting the image into an electrical signal. Venous images and blood flow information may be obtained from the captured images, including filtering algorithms for veins having a thickness of less than 1 mm. The measured blood flow flow information is determined whether or not it is a living body by verifying whether it is within a range of pre-registered biometric information, and compares the vein pattern extracted through the laser speckle image method with a pre-registered vein pattern. Biometric authentication with improved security is possible.

User authentication may be performed by comparing an image having improved biometric determination and extracted accuracy with a previously stored vein image pattern. In addition, the blood flow distribution information through the extracted speckle imaging technique can be used to verify the validity of the vein pattern.

As described above, specific parts of the present disclosure have been described in detail, and for those skilled in the art, these specific descriptions are merely preferred embodiments, and the scope of the present disclosure is not limited thereto. Will be obvious. Thus, the substantial scope of the present invention will be defined by the appended claims and their equivalents.

1: light source device
2: light source optical system
3: light source control device
4: image sensor
5: sensor optical system
6: Speckle Analysis Device
7: video output device

Claims (7)

(a) a sensor for recognizing the existence of the accommodation space for receiving the authentication target site in a non-contact manner and the authentication target site in the accommodation space;
(b) a laser irradiation unit for irradiating a laser to the authentication target site;
(c) an image generator for converting light reflected by the laser irradiation into an image through a laser speckle imaging method;
(d) an imaging sensor to capture the converted image;
(e) a biometric determination unit configured to determine whether the information value input from the imaging sensor is within a pre-registered biometric value range and determine whether the authentication target site is a living body;
(f) an image processor extracting vein pattern information from the captured image through a noise filtering algorithm and displaying the vein pattern as an image having improved accuracy; And
(g) a biometric authentication unit configured to perform user authentication by comparing the vein pattern image previously stored in a memory with an image of improved biometric determination and accuracy; Biometric authentication device comprising a.
The biometric authentication device according to claim 1, wherein the biometric authentication target portion is a back of the hand and / or a palm.
The biometric device according to claim 1, wherein the laser is a near infrared ray having a wavelength of 700 nm to 1000 nm with high coherence.
The biometric device of claim 1, wherein the imaging sensor comprises a CCD camera having a near infrared filter.
The biometric device of claim 1, wherein the image processor extracts pattern information of a vein having a thickness of 1 mm or more and 2 mm to represent an image having improved accuracy of the vein pattern.
The biometric authentication method according to claim 1, wherein the image processing unit is a computing device selected from the group comprising a tablet PC, an embedded system, a PDA, a mobile phone, a computer, a microprocessor, a microcomputer, a microcontroller, and other related devices. Device.
(a) converting the light reflected by the laser irradiation to the biometric authentication target area into an image through a laser speckle imaging method;
(b) capturing the converted image;
(c) determining whether the authentication target site is biometric based on the input blood flow information;
(d) extracting vein pattern information from the captured image through a noise filtering algorithm to represent the vein pattern with improved accuracy; And
(e) performing user authentication by comparing the vein pattern image previously stored in the memory with the image of which the biometric determination and accuracy are improved; Biometric authentication method using the biometric authentication device of any one of claims 1 to 6.

Images