KR101606351B1 - Method and apparatus for discriminating imitation-body - Google Patents

Abstract

The present invention relates to a method and an apparatus for discriminating an imitation living body, the method comprising: generating motion reference information by grasping a face region and an eye region in an initial image of a subject; A step of recognizing a facial region and an eye region to generate motion comparison information; comparing the motion reference information and the motion comparison information to determine a facial motion characteristic of the subject; Or not.

Description

METHOD AND APPARATUS FOR DISCRIMINATING IMITATION-BODY [0002]

The present invention relates to a method and apparatus for dummy living body identification. And more particularly, to a method and apparatus for identifying a living body based on a living body by analyzing a photographed image of a subject.

As well as the development of the information and the Internet, the scale of the industry related to the online electronic commerce and the identity verification has been enlarged and the demand of the accurate personal authentication has become bigger, and as a means of authenticating it, a PIN (Personal Identification Number ) Or the limit of the password method reached saturation.

On the other hand, user authentication through biometrics is widely used in embedded systems or personal computers such as ATMs (Automated Teller Machines) and the like, In recent years, the use of mobile devices has been expanding as technology advances.

The most important characteristic of biometric information, including fingerprints, is that there is no risk for forgetting such as a password, and there is a unique advantage that the authentication can not be performed unless it is the person himself.

However, since such authentication biometric information is converted into digital data and used in a personal computer, a portable device, a server, etc., if such a device hacking accident occurs, biometric information for authentication can be leaked. The information could be used for malicious biometric authentication using imitation biometrics.

Therefore, at the time of biometric authentication, it is necessary to judge whether the biometric information for authentication is a living body or a living body of a dummy body.

On the other hand, biometrics authentication methods vary widely, such as a method of using fingerprint, a method of using face, a method of using iris, and a method of using blood vessel. However, according to the related art, techniques for discriminating phonetic fingerprint are mainly used. And it was not judged whether or not the living body was a dummy living body.

Korean Registered Patent No. 1179559, registered on August 29, 2012.

According to the embodiment of the present invention, it is possible to use a face image to determine whether a living body is a living body, regardless of the type of the biometric authentication system, A method and apparatus for identifying a living body are provided.

The problems to be solved by the present invention are not limited to those mentioned above, and another problem to be solved can be clearly understood by those skilled in the art from the following description.

According to a first aspect of the present invention, there is provided a method of discriminating a living body from a subject, comprising: generating motion reference information by grasping a face region and an eye region in an initial image of a subject; The method comprising the steps of: generating motion comparison information by recognizing an area of a subject; comparing the motion reference information and the motion comparison information to obtain a face motion characteristic of the subject; And determining whether the subject is a living subject.

Wherein the motion reference information and the motion comparison information include face rotation state information and pupil rotation state information, and wherein the step of determining whether the imitation living body includes the face rotation state information and the pupil rotation state The subject can be determined to be a dummy living body when the facial motion characteristic according to the comparison result of information does not correspond to the vestibular reflex motion.

In the imitation living body determination method, the face rotation state information may indicate a relative placement angle between the position of the nose or mouth and the position of the eye.

A dummy living body identifying apparatus according to a second aspect of the present invention is a dummy living body identifying apparatus for detecting a face region and an eye region in an initial image of a subject to generate motion reference information, A movement detecting unit for detecting a face movement characteristic of the subject by comparing the motion reference information and the motion comparison information to generate motion comparison information; And an imitation judging section for judging whether the imitation living body is a living body or not.

Wherein the motion reference information and the motion comparison information include face rotation state information and eye pupil rotation state information, and the imitation determination unit determines whether or not the face rotation state information and the pupil rotation state information It is possible to determine that the subject is a dummy living body when the facial motion characteristic according to the present invention does not correspond to the vestibular ocular reflex motion.

In the imitation living body determination method, the face rotation state information may indicate a relative placement angle between the position of the nose or mouth and the position of the eye.

According to the embodiment of the present invention, it is possible to accurately determine whether the imitation living body is using the face by paying attention to the fact that the face is located at least close to the living body irrespective of the type of the biometric authentication system, do.

The imitation biometric identification technology provided in accordance with the embodiment of the present invention can be applied to all biometric authentication methods such as a method using a fingerprint, a method using a face, a method using an iris, a method using a blood vessel Can be applied.

FIG. 1 is a block diagram of a dummy biometric identification device according to an embodiment of the present invention.
FIG. 2 is a flowchart for explaining a dummy living body identification method according to an embodiment of the present invention.
3 is a view showing a process of changing a relative arrangement angle between the position of the nose and the position of the eye and the relative arrangement angle between the position of the mouth and the position of the eye according to the rotation of the face.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims.

In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. The following terms are defined in consideration of the functions in the embodiments of the present invention, which may vary depending on the intention of the user, the intention or the custom of the operator. Therefore, the definition should be based on the contents throughout this specification.

A method and apparatus for dummy living body identification according to an embodiment of the present invention determines a living body to be a living body for a subject that normally performs vestibular ocular reflex movement through a Vestibulo-Ocular Reflex (VOR) test on a subject, For a subject not performing a reflex motion, it is judged to be a dummy living body. The vestibular reflex is a representative biomechanical mechanism that must be implemented in the visual action of living things. It is a reflective eye movement that stabilizes the image of the body even during the movement of the body. The vestibular reflex is to keep the movement of the eye automatically so that the image of the outside world is kept in the retina as long as possible, causing the eye movement in the direction opposite to the direction of movement of the head when the human head moves.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a configuration diagram of a dummy living body identifying apparatus according to an embodiment of the present invention; FIG.

As shown in the figure, the dummy biometric identification apparatus 100 according to the embodiment includes a motion information generation unit 110, a motion detection unit 120, a dummy determination unit 130, and the like.

The motion information generation unit 110 generates motion reference information by recognizing a face area and an eye area in an initial image of a subject and grasps a face area and an eye area in a next image of the subject by time flow to generate motion comparison information do.

The motion information generation unit 110 analyzes the photographed image over a plurality of times based on the subject and generates face rotation state information and pupil rotation state information based on the face rotation motion of the head. Here, the face rotation state information may be expressed as a relative arrangement angle between the position of the nose and the position of the eyes, or may be expressed as a relative arrangement angle between the position of the mouth and the position of the eyes.

The motion detecting unit 120 compares the motion reference information generated by the motion information generating unit 110 with the motion comparison information to determine the facial motion characteristics of the subject. The motion detecting unit 120 may compute the face motion characteristics of the subject by comparing the face rotation state information and the pupil rotation state information included in the motion comparison information and the motion comparison information.

The dummy determination unit 130 determines whether the dummy body is a living body based on the face movement characteristic detected by the motion detection unit 120. When the face motion characteristic according to the result of comparing the motion state information and the motion state information and the eye motion state information included in the motion comparison information does not correspond to the vestibular reflex motion, It is judged as a living body.

FIG. 2 is a flowchart for explaining a dummy living body identification method according to an embodiment of the present invention.

As shown in the figure, the dummy biological body determination method according to the embodiment includes step S201 of generating motion reference information by grasping a face region and an eye region in an initial image of a subject. Here, the face rotated state information and the pupil rotation state information can be generated by analyzing the image photographed over a plurality of times in accordance with the time of the subject, and by rotating the face of the head. For example, the face rotation state information may be expressed as a relative placement angle between the position of the nose and the position of the eye, or may be expressed as a relative placement angle between the position of the mouth and the position of the eye.

The method further includes a step S203 of generating motion comparison information by grasping the face region and the eye region in the next image of the subject by the time flow.

In addition, the step of comparing the generated motion reference information and the motion comparison information to determine the facial motion characteristic of the subject (S205). Here, it is possible to compare the face rotation state information and the pupil rotation state information included in the motion reference information and the motion comparison information.

Subsequently, steps S207 to S211 are performed to determine whether the subject is a dummy living body according to the detected facial motion characteristics. Here, if the facial motion characteristic according to the comparison result of the face rotation state information and the pupil rotation state information included in the motion reference information and the motion comparison information does not correspond to the vestibular reflex motion, the subject can be determined as the imitation living body have.

Hereinafter, a dummy biometric identification process performed by the dummy biometric identification device according to the embodiment of the present invention will be described in detail with reference to FIG. 1 to FIG.

In order to discriminate a dummy living body according to an embodiment of the present invention, specific action tips related to image shooting are promised beforehand between a subject to be subjected to biometric identification, that is, a person and a dummy living body identification device 100. For example, a person should cooperate to photograph a front face of a face by looking at a camera that can be included in the motion information generation unit 110, perform a face rotation motion to turn the head while looking at the camera, Should be able to photograph the side of the face.

When a person watches the camera in a state where such an action is promised in advance, the motion information generating unit 110 acquires an initial image with respect to the subject, grasps the face region and the eye region in the initial image of the subject, And generates reference information (S201).

Here, the motion information generation unit 110 generates the face rotation state information and the pupil rotation state information, and the face rotation state information is represented by a relative placement angle between the position of the nose and the position of the eye, As shown in FIG.

For this purpose, the motion information generation unit 110 analyzes the initial image, detects the position of one eye, which is one of the face elements, detects the center of the nose or mouth as a feature point, estimates the face region based on the positions of both eyes can do. For example, the position of both eyes can be detected using a known adaboost algorithm or the like. In order to generate the pupil rotation state information, image binarization of the eye region including the pupil can be performed to confirm in which direction the pupil is biased.

Thereafter, when a person performs a face rotation movement while turning his / her head while looking at the camera according to the promised action, the motion information generation unit 110 photographs the side of the face and acquires the next image according to the time course of the subject .

Next, the motion information generating unit 110 grasps the face region and the eye region in the next image of the subject to generate motion comparison information. Here, the motion information generation unit 110 may generate the face rotation state information and the pupil rotation state information by analyzing the next image of the subject based on the same method or principle as that at the time of generation of the motion reference information (S203).

Next, the motion recognition unit 120 compares the motion reference information generated by the motion information generation unit 110 with the motion comparison information, grasps the facial motion characteristics of the subject based on the comparison result, And provides the facial motion characteristic of the facial motion estimation unit 130 to the facial motion estimation unit 130. [ Here, the motion detecting unit 120 compares the face rotation state information and the pupil rotation state information included in the motion reference information, the motion comparison information, and determines whether the subject rotates the face in one direction, (S205).

Then, the dummy determination unit 130 determines whether the dummy body is a living body based on the face movement characteristics captured by the motion recognition unit 120 (S207). When the facial motion characteristic corresponds to the vestibular reflex motion, the imitation determination unit 130 determines that the subject is a living body (S209). If the facial motion characteristic does not correspond to the vestibular reflex motion, (S211). That is, when the gazing direction of the eye is grasped in the direction opposite to the face rotation direction of the subject, the phantom determination unit 130 determines that the subject is a living body by judging that the vestibular ocular reflex motion is performed properly.

3 (b) is an initial image of a subject obtained by the motion information generating unit 110, and FIG. 3 (b) is an initial view of the subject obtained by the motion information generating unit 110. Referring to FIG. 3, (a) and (c) are the next images of the subject acquired by the motion information generation unit 110. [ The pupil center positions A, B, A ', B', A ", and B " Then, the position of the nose tip or the lip center position is detected as the feature point. That is, the positions C, C ', and C "of the nose tip are detected as minutiae or the lip center positions D, D', D" are detected as minutiae.

3 (b), the difference between ∠BAC and ∠ABC is small in the triangle connecting the center of both eyes and the center of the nose. On the other hand, in Fig. 3 (a), the difference between ∠B'A'C 'and ∠A'B'C' is large. This is also true in FIG. 3 (c), and the same result can be obtained when the lip center position is detected as the feature point.

Therefore, the motion recognition unit 120 can compare the sizes of the two internal angles and determine that the face has rotated toward a larger angle change relative to the initial position.

Although the embodiments for grasping the face rotation direction have been described above, the face rotation direction and the pupil rotation direction can be detected and grasped based on other methods or principles. For example, image difference values between consecutive frames of a camera may be utilized, and eye movement and face movement may be grasped by using eye tracking technology.

Combinations of the steps of each flowchart attached to the present invention may be performed by computer program instructions. These computer program instructions may be loaded into a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus so that the instructions, which are executed via a processor of a computer or other programmable data processing apparatus, Lt; / RTI > These computer program instructions may also be stored in a computer usable or computer readable memory capable of directing a computer or other programmable data processing apparatus to implement the functionality in a particular manner so that the computer usable or computer readable memory It is also possible to produce manufacturing items that contain instruction means for performing the functions described in each step of the flowchart. Computer program instructions may also be stored on a computer or other programmable data processing equipment so that a series of operating steps may be performed on a computer or other programmable data processing equipment to create a computer- It is also possible for the instructions to perform the processing equipment to provide steps for executing the functions described in each step of the flowchart.

In addition, each step may represent a module, segment, or portion of code that includes one or more executable instructions for executing the specified logical function (s). It should also be noted that in some alternative embodiments, the functions mentioned in the steps may occur out of order. For example, the two steps shown in succession may in fact be performed substantially concurrently, or the steps may sometimes be performed in reverse order according to the corresponding function.

The foregoing description is merely illustrative of the technical idea of the present invention and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

100: Imitation biometric discriminator
110: a motion information generating unit
120:
130: Mojo board

Claims (6)

Generating motion reference information including face rotation state information and pupil rotation state information by grasping a face region and an eye region in an initial image of a subject,
Generating motion comparison information including the face rotation state information and the pupil rotation state information by grasping the face region and the eye region in the next image of the subject by time flow,
Comparing the motion reference information and the face rotation state information included in the motion comparison information and comparing the pupil rotation state information to obtain a face motion characteristic of the subject,
And judging the subject to be a dummy living body when the detected facial motion characteristic does not correspond to the vestibular ocular reflex motion .
delete The method according to claim 1,
Wherein the face rotation state information indicates a relative placement angle between a position of the nose or mouth and a position of the eye.
The face region and the eye region are grasped in the initial image of the subject to generate motion reference information including the face rotation state information and the pupil rotation state information and the face region and the eye region are grasped in the next image of the subject by the time flow A motion information generating unit for generating motion comparison information including face rotation state information and eye pupil rotation state information ,
Comparing the face rotation state information included in the motion reference information and the motion comparison information and comparing the pupil rotation state information to obtain a face movement characteristic of the subject;
And an imitation judging unit for judging whether or not the subject is a dummy living body when the detected face movement characteristic does not correspond to the vestibular ocular reflex movement .
delete 5. The method of claim 4,
Wherein the face rotation state information indicates a relative placement angle between a position of the nose or mouth and a position of the eye.

Images