KR20160040037A - The object authentication method and apparatus by the short-range video detection of multiple structure - Google Patents

Abstract

The present invention relates to an object authentication device by short-range video detection of a multi-structure and a method thereof. The device comprises: a distance sensor (712); a camera (711) for recognizing the face of a person who passes; a face processing unit (110); a fingerprint recognizing sensor (713) for recognizing a fingerprint of the person who passes; a fingerprint processing unit (210); a speed improving unit (211) for increasing a matching speed; a fingerprint database (612); a face database (611); a network connecting unit (510) connectable to an external terminal through a wired/wireless network; and a control unit (610) for controlling a lock device of a door to be opened and closed through authentication. Therefore, the device increases a processing speed and accuracy when an object is authenticated.

Description

[0001] The present invention relates to an object authentication method and apparatus for short-range video detection,

[0001] The present invention relates to an apparatus and method for authenticating an object by means of a multi-structure near-field image detection, more specifically, a method for identifying a fingerprint and a face recognition in a double manner, The present invention relates to an apparatus and method for authenticating an object by a multi-structure short range image sensing, characterized in that a user is identified by applying and analyzing a recognition algorithm, and then fingerprints are recognized to improve a processing speed and accuracy.

As the aging and personalization of the society are progressing, interest in the security field is increasing. In the digital era, identification methods today include passwords, smart cards, HW tokens, personal identification cards, personal identification cards, and so on. Biometrics and so on. However, people do not like to carry cards or keep track of passwords that are only a few dozen, so biometric technology has provided consumers with a safer and more convenient lifestyle.

User authentication, which is the basis of biometric technology, is growing in importance with the expansion of non-face-to-face electronic transactions. Biometric technology has been selected as one of the 20 most promising technologies of the 21st century by Gartner Group and MIT. Compared to the same period of last year.

Worldwide, since June 2008, ICAO has applied biometric-based immigration control systems to more than 45 countries in accordance with the recommendations of the International Civil Aviation Organization (ICAO), improved national security by detecting terrorist suspects, By preventing identification of ID cards by using information, criminal activities by dual IDs were blocked, and projects related to national security were carried out. In addition, biometric technology is applied to access control of major facilities, network infrastructure, and access to important systems. However, it is widely used in various applications such as ensuring the safety of infrastructure. Basically, registration / enrollment), identification (verification, one-to-one comparison), and identification (identification, one-to-many comparison) that finds users in the database. Authentication and identification are selectively used depending on the application area.

In the biometric system, fingerprint recognition is performed using the fingerprint obtained from the original image of the user to be authenticated and the fingerprint feature information of the registered user.

At this time, the feature information of the fingerprint uses an algorithm for extracting and storing the positions and angles of the feature points including the bifurcation points and the end points of the fingerprint bumps.

Fingerprint recognition can reduce the device cost by miniaturizing the recognizer as compared with other biometric products, and it accounts for more than 33% of the biometric products currently marketed with high reliability. However, 5% of the population is not able to register fingerprints because fingerprints are faint, worn out or lost due to lack of fingers, births, or late-effects. When fingerprints are captured in the scanner, And a recognizer which is not equipped with a reinforcement algorithm corresponding to a counterfeit fingerprint has a disadvantage in that it is weak in modulation.

In addition, these biometric systems have a lot of recognition errors, and there is a limit to use as a perfect security system because there is a possibility of rejecting a correct accessor or entering an unregistered accessor. In addition, face recognition requires a considerable amount of time in the process of comparing data in order to increase the amount of data and increase the accuracy of the recognition.

KR 10-2012-0122587 KR 10-2000-0026143 KR 10-2000-0060312

SUMMARY OF THE INVENTION The present invention has been made in order to solve the above-mentioned problems of the related art, and it is an object of the present invention to provide a face recognition system, in which a camera is operated by a distance sensor when a person approaches, And then authenticating the user through a method of recognizing the fingerprint. That is, the object of the present invention is to improve the processing speed and accuracy of the object authentication.

In order to accomplish the object of the present invention, there is provided a method for identifying a person using a face recognition and a fingerprint recognition method together with a multibody recognition method Specifically, a distance sensor 712 that operates when a person approaches; A camera 711 for recognizing the face of the passenger; A face processing unit 110 for extracting a feature from the facial image signal recognized through the camera 711 and converting the feature into data; A fingerprint recognition sensor 713 for recognizing the fingerprint of the passenger; A fingerprint processing unit 210 for extracting features from fingerprints recognized through the fingerprint input sensor 713 and converting the extracted features into data; A speed increasing unit 211 for increasing a matching speed in recognizing the digitized fingerprint; A fingerprint database 612 storing IDs of persons authorized to enter and leave and corresponding fingerprint data; A face database 611 in which an ID shared with the fingerprint database 612 and face data of the corresponding person are stored; A network connection unit 510 capable of connecting to an external terminal through a wired / wireless network; While the camera operated by the distance sensor 712 is approaching the passenger, the face data of the passenger inputted from the face processing unit 110 is stored in the face data stored in the face database 611 Comparing the fingerprint data of the visitor inputted from the fingerprint processing unit 210 with the corresponding fingerprint data of the ID of the face of the fingerprint database 612 to authenticate the face, A control unit 610 for transmitting the fingerprint of a passenger to the terminal 410 of the administrator through the wired / wireless network using the network connection unit 510, receiving and authenticating the authentication signal from the administrator, and controlling the door lock unit to be opened and closed; And a control unit.

As described above, in the case of using the apparatus and method for authenticating a short-range image by multi-structured image according to the present invention, it is possible to control whether or not a passenger enters or exits through two-step authentication such as face authentication and fingerprint authentication, It is possible to maintain high security by storing facial images and fingerprints in the case of a person who has not been recognized as a disadvantage of fingerprint recognition, or a person who has been worn out or lost due to a later cause, Can be improved.

In addition, when a distance sensor is provided in the face processing unit, if the user is detected within a certain radius of the object authentication apparatus, the automatic operation can be performed to shorten the time required for face recognition considerably.

In addition, the unidirectional matching is basically used in the fingerprint processing unit, but a separate threshold concept is introduced in addition to the threshold used in the matching success determination, so that only when the matching score exists between the existing threshold value and the new threshold value, And a processing speed of 1: 1 is provided even when a speed increasing unit for greatly reducing the average time per matching is provided and the processing is performed in the 1: N manner, thereby reducing the time required for fingerprint recognition Can be obtained.

Also, by connecting the terminal of the administrator through the wired / wireless network and transmitting the facial image of the visitor, and receiving the authentication signal from the manager, the administrator can control the access even from a remote place, and even if not the registered person, Access to a person who has a fingerprint can be granted, and even when an error occurs in the authentication process or fingerprint recognition can not be performed due to a birth reason or a later cause, have.

FIG. 1 is a block diagram of an object authentication apparatus according to the present invention.
2 is a schematic flow diagram of an object authentication method according to the present invention;
3 is a detailed flowchart of an object authentication method according to the present invention;
FIG. 4 is a flowchart illustrating another embodiment of an authentication step in an object authentication method according to another embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings.

FIG. 1 is a block diagram of an object authentication apparatus using a multi-structure short-range image detection according to the present invention.

The present invention relates to an apparatus and method for authenticating an object by combining a face recognition method and a fingerprint recognition method, and more particularly, to an apparatus and method for authenticating an object by combining a face recognition method and a fingerprint recognition method, including a distance sensor 712 operated when a visitor approaches within a certain radius of a multi- A camera 711 for recognizing the face of the passenger; A face processing unit 110 for extracting a feature from the facial image signal recognized through the camera 711 and converting the feature into data; A fingerprint recognition sensor 713 for recognizing the fingerprint of the passenger; A fingerprint processing unit 210 for extracting features from fingerprints recognized through the fingerprint recognition sensor 713 and for data generation; a speed increasing unit 211 for increasing a matching speed in recognizing the digitized fingerprint; A fingerprint database 612 storing IDs of persons authorized to enter and leave and corresponding fingerprint data; A face database 611 in which an ID shared with the fingerprint database 612 and face data of the corresponding person are stored; A network connection unit 510 capable of connecting to an external terminal through a wired / wireless network; While the camera operated by the distance sensor 712 is approaching the passenger, the face data of the passenger inputted from the face processing unit 110 is stored in the face data stored in the face database 611 Comparing the fingerprint data of the visitor inputted from the fingerprint processing unit 210 with the corresponding fingerprint data of the ID of the face of the fingerprint database 612 to authenticate the face, A control unit 610 for transmitting a fingerprint of a passenger to the administrator terminal 410 via the wired / wireless network using the network connection unit 510 and receiving and authenticating an authentication signal from the manager to control opening and closing of a lock device of the door .

The flow of the object authentication method using the multi-structure object authentication apparatus by the near vision image detection is schematically shown in FIG.

2, the method for authenticating an object according to the present invention includes the steps of recognizing an entrance (S100), an entrance face authentication step S200, a fingerprint recognition step S300, a fingerprint authentication step S400, an unlocking step S500 ).

That is, when the accessing party approaches the multi-layered local image sensing apparatus, the passenger recognizes the passenger by the distance sensor 712, extracts the face image of the passenger, and if the face is authenticated, receives the fingerprint to determine whether the fingerprint is registered.

As described above, the object authentication method according to the present invention proposes higher security by recognizing an accessing person through face authentication and fingerprint, and can prevent a fingerprint recognized as a disadvantage of fingerprint recognition from being worn or missing It is possible to improve the inconvenience when the user authenticates the object only by the fingerprint recognition method.

3 is a flow chart for explaining the access control method according to the present invention in more detail.

As shown in FIG. 3, the object authentication method according to the present invention includes a passenger access recognition step (S110) for recognizing a passenger by a distance sensor (712) when a passenger approaches the object authentication device, a passenger access recognition step (S210) for recognizing the face of the passenger recognized by the face recognition step (S210), recognizing the face of the passenger recognized by the face recognition step (S210) A facial database registration step S230 for determining whether the facial image of the visitor is registered in the face database 611 or a facial database registration step S230 for authenticating the face A fingerprint recognition step (S300) of recognizing a fingerprint through the fingerprint recognition sensor (713), a feature of the fingerprint recognized from the fingerprint recognition step (S300) A fingerprint data registration step S420 of determining whether the recognized fingerprint is a fingerprint stored in the fingerprint data, a fingerprint data registration determination step S420 of comparing the fingerprint data with the fingerprint data stored in the base 612, (S510) for releasing the locking device when the authentication is completed.

If it is determined that the face or the fingerprint is not the registered face or fingerprint in the face database registration determination step S230 or the fingerprint data registration determination step S420, the authentication step S210, S220, S230, S300, (S600) for storing the recognized face of the passenger recognized through the camera 711 in steps S410, S410, and S420, and a recognized fingerprint storage step S610 for storing the recognized fingerprint.

Here, the fingerprint database 612 basically uses unidirectional matching in recognizing the digitized fingerprint, introduces a separate threshold concept other than the threshold used in the matching success determination, And the biometric matching is performed only when there is a value between the values, and the average time per matching is greatly reduced. Therefore, even when 1: N processing is performed in a 1: N manner, And the time required for fingerprint recognition is shortened.

4 is a flowchart illustrating another embodiment of the authentication step in the object authentication method according to the present invention.

4, the step of authenticating the face of the passenger includes the steps of acquiring the face image of the passenger through the camera 711 (S700), the controller 610 acquiring the face image of the passenger through the wired / (S710), which is connected to the administrator terminal (410) through the administrator terminal connection step (S710), and the face image acquired from the visitor face image acquisition step (S700) An authentication signal input step S730 for inputting an authentication signal from the administrator terminal 410, an authentication signal inputted from the authentication signal input step S730, (S740); a door lock release step (S750) of releasing a locking device of the door when the authentication is determined by the authentication determination step (S740) For unauthorized chulipja in step (S740) comprises a face unverified storing step (S760) for storing the face image obtained from the face image obtaining step (S700).

Although FIG. 4 illustrates an example of the face authentication step, if the authentication is not performed at the fingerprint authentication step as well as the face authentication step, the user can be directly authenticated to the administrator through the wired / wireless network.

711: Camera
712: Distance sensor
713: Fingerprint sensor
210: Fingerprint processor
211: speed increasing section
611: face database
612: fingerprint database
410: Administrator terminal
510: network connection
610:
S100: Person Identification Step
S200: passenger face authentication step
S300: fingerprint recognition step
S400: Fingerprint authentication step
S500: Unlock phase
S110: Entrance access recognition step
S210: passenger face recognition step
S220: Face database search step
S230: Face database registration determination step
S410: Fingerprint data comparison step
S420: Fingerprint data registration determination step
S510: Unlocking step

Claims (11)

1. An apparatus for authenticating an object by a near-field image detection of a multi-
A distance sensor 712 that operates when a person approaches within a certain radius of the multi-structure short range image sensing device;
A camera 711 for recognizing the face of the passenger;
A facial processing unit 110 for extracting features from facial image signals recognized through the camera and converting the extracted features into data;
A fingerprint recognition sensor 713 for recognizing the fingerprint of the passenger;
A fingerprint processing unit 210 for extracting features from fingerprints recognized through the fingerprint recognition sensor and converting the extracted features into data;
A speed increasing unit 211 for increasing a matching speed in recognizing the digitized fingerprint;
A fingerprint database 612 storing IDs of persons authorized to enter and leave and corresponding fingerprint data;
A face database 611 in which an ID shared with the fingerprint database and face data of the corresponding person are stored;
A network connection unit 510 capable of connecting to an external terminal through a wired / wireless network;
While the camera operated by the distance sensor 712 is approaching the passenger, the face data of the passenger inputted from the face processing unit 110 is stored in the face data stored in the face database 611 Comparing the fingerprint data of the visitor inputted from the fingerprint processing unit 210 with the corresponding fingerprint data of the ID of the face of the fingerprint database 612 to authenticate the face, A control unit 610 for transmitting a fingerprint of a passenger to the administrator terminal 410 via the wired / wireless network using the network connection unit 510 and receiving and authenticating an authentication signal from the manager to control opening and closing of a lock device of the door Wherein the object authentication device is a multi-structure object authentication device. In an object authentication method using a multi-structure short-range image detection,
An entrance recognizing step (SlOO) in which a camera is automatically operated when a user approaches the object authentication apparatus a certain distance;
A passenger face authentication step (S200) of judging whether the face of the passenger is a registered face image by comparing the face database (611) stored beforehand with a match;
A fingerprint recognition step (S300) of recognizing the fingerprint of the visitor;
A fingerprint authentication step (S400) of comparing the fingerprint data recognized in the fingerprint recognition step (S300) with the fingerprint data stored in the fingerprint database (612) to determine whether the fingerprint of the user is a registered fingerprint;
And a lock release step (S500) of releasing a locking device of a door when it is determined that the fingerprint is registered in the fingerprint authentication step (S400). 3. The method of claim 2,
The passenger face authentication step (S200)
A face recognition step (S210) of recognizing the face of the passenger recognized in the passenger access recognition step (S110) through the camera; A face database searching step (S220) of searching the face database (611) by extracting the face or the recognized characteristic recognized by the entering face recognition step (S210); And a face database registration determination step (S230) of determining whether the face image of the user is a face registered in the face database (611). 3. The method of claim 2,
The fingerprint authentication step (S400)
A fingerprint data comparison step (S410) of extracting the recognized fingerprint characteristic from the fingerprint recognition step (S310) and comparing the fingerprint data with the fingerprint data stored in the fingerprint database (612); Further comprising a fingerprint data registration determination step (S420) of determining whether the recognized fingerprint is a fingerprint stored in the fingerprint data (S420). The method according to claim 1,
And the distance sensor 712 allows the camera to automatically operate when the user approaches a distance of the image sensing device. The method according to claim 1,
And the camera 711 is automatically operated by the distance sensor 712 to operate the face database 611 to acquire a facial image of a passenger. . The method according to claim 1,
The face processing unit 110 further determines whether or not the contour structure of the face inputted to the camera 711 and the face data of the face database 611 coincide with each other. Object authentication device. The method according to claim 1,
Wherein the fingerprint recognition sensor 713 is a sensor for recognizing the fingerprint recognition of the passenger when the face of the passenger coincides with the face database 611 stored in advance. Object authentication device. The method according to claim 1,
The speed enhancement unit 211 basically uses unidirectional matching in recognizing the digitized fingerprint, introduces a separate threshold concept in addition to the threshold used in the matching success determination, And a processing speed in a 1: 1 manner even in the case of processing in a 1: N manner by a matching algorithm which performs bidirectional matching only when there is a value between the values and significantly reduces an average time per matching. Object Identification Device by Near - field Image Detection of Structure. 3. The method of claim 2,
If the face or the fingerprint is not a face or a fingerprint registered in the passenger face authentication step S200 or the fingerprint authentication step S400, the entrance party recognized in the entrance party face recognition step S100 or the fingerprint recognition step S300 (S760) for storing a face image or a fingerprint of the face image of the object to be authenticated. The method according to claim 1,
The network connection unit 510 transmits the face image of the passenger inputted through the camera 711 to the administrator terminal 710 in the case where the passenger is not authenticated in the passenger recognition step S100 or the fingerprint recognition step S300. And an authentication signal indicating permission or denial of entry / exit from the manager can be received and authenticated.

Images