TWI585607B - Eye movement traces authentication and facial recognition system, methods, computer readable system, and computer program product - Google Patents

Description

Eye line trajectory authentication system, method and computer readable recording medium combined with face authentication And computer program products

The present invention relates to a line of sight trajectory authentication system and method thereof, and more particularly to a line of sight trajectory authentication system and method thereof combined with face authentication or autonomous will confirmation.

Authentication means that the user's identity is checked through certain means, and the user's authority is determined by checking the user's identity. In the past, the technology for identity verification was mostly verified by a password, and the combination of numbers or graphics was used to assign the password combination to the corresponding user. The combination of passwords may correspond to a user's rights including obtaining the same item, obtaining a piece of information, or access rights, and the like.

The disadvantage of password verification is that it only confirms the user's identity through the user's mental power. Therefore, if the password is leaked, even if the person who entered the password is not the user, the device will not check the identity of the user, which will lead to identity. Verification ability produces flaws. And because of such problems, password theft is endless, such as obtaining the user's password through the fingerprint remaining on the keyboard, or through the side. Record the way to record the password typed by the user, and so on.

Therefore, in order to solve the above problem, in the past, the concept of tracking the user's eyes to input the user's password has been proposed. By tracking the device for measuring the position of the eyeball and the information on the movement of the eyeball, the user can input a plurality of passwords displayed on the screen by looking at different positions on the screen, thereby preventing others from stealing the password through illegal means. However, by entering the password through the gaze direction, it is still possible for the side recording device to record the user's eye movement trajectory, thereby deriving the possibility of the password on the screen by the eye movement mode, and therefore, how to prevent the user from being recorded by the side recording device. The situation that causes the password to be leaked is a problem that must be solved by those skilled in the art.

SUMMARY OF THE INVENTION An object of the present invention is to solve the problem that in the past, when a password is input by eye movement, the eye track may be recorded by the skimming device, thereby causing the password to be stolen.

In order to solve the above problems, the present invention provides a visual line trajectory authentication system including an image capturing device, a display device, and a control unit. The image capturing device is used to capture a user image. The display device provides a display interface for the user to look at. The control unit includes an eye movement analysis module, a line of sight trajectory unlocking module, and an autonomous will confirmation module. The eye movement analysis module obtains the gaze direction of the user according to the eye image of the user image. The gaze track unlocking module records the gaze direction of the user to obtain a trajectory trajectory information, compares the gaze trajectory information with the trajectory data in the database, and generates a control signal to transmit to the comparison result when the comparison result is met. Identity verification device. The autonomic confirmation module provides a graphical interface or a hidden interface for the user to gaze at the input, and the graphical interface provides a smart query function. And according to the identity data input by the user in advance, after generating the corresponding inquiry content, obtaining the eye movement or the gaze direction of the user, whether the option selected by the user's eye movement or gaze direction is in advance The set options are the same, and depending on the result of the inquiry content, it is determined whether the user operates under his or her will.

Another object of the present invention is to provide a line-of-sight track authentication system incorporating face authentication, comprising an image capture device, a display device, a control unit, and an autonomous confirmation module. The image capturing device is used to capture a user image. The display device provides a display interface for the user to look at. The control unit includes a facial feature recognition module, an eye movement analysis module, and a line of sight trajectory unlocking module. The facial feature recognition module acquires the user image, and extracts the facial feature from the user image, and compares the facial feature with the facial data in the database to verify the user identity. The eye movement analysis module obtains the gaze direction of the user according to the eye image of the user image. The gaze track unlocking module records the gaze direction of the user to obtain a trajectory trajectory information, compares the gaze trajectory information with the trajectory data in the database, and generates a control signal to transmit to the comparison result when the comparison result is met. Identity verification device. The self-willing confirmation module provides a graphical interface or a hidden interface for the user to gaze into the input. The graphical interface provides a smart inquiry function, and according to the user-entered identity data, the corresponding inquiry content is generated and obtained. Whether the user's eye movement or gaze direction selects an option selected by the user's eye movement or gaze direction to be the same as a preset option, and determines whether the user operates under the autonomous will according to the result of the inquiry content. .

Further, the control unit includes a graphic generation module. The graphics generation module provides a line of sight trajectory input interface for the user to look at the operation to input the line of sight trajectory information according to the gaze direction of the user.

Further, the graphic generation module provides an auxiliary graphic for the user to look at in the visual line trajectory input interface, so that the user inputs the visual line trajectory information according to the auxiliary graphic.

Further, the facial feature is an iris image of an iris region in the eye image of the user.

Further, the facial feature is a facial feature image in the facial image of the user.

Further, the facial feature is a blood vessel distribution image of the scleral region in the eye image of the user.

A further object of the present invention is to provide a method for authenticating a gaze track, comprising: capturing a user image; extracting an eye image through the user image, and obtaining a gaze direction of the user from the eye image; recording a gaze direction of the user Obtaining a line of sight information; comparing the line of sight information with the track data in the database, and generating a control signal to the identity verification device when the comparison result is met; and providing a graphical interface or a hidden interface For the user to look at the input, the graphical interface provides a smart inquiry function, and according to the identity data previously input by the user, after generating the corresponding inquiry content, obtaining the eye movement or the gaze direction of the user, by using the Whether the selected option of the user's eye movement or gaze direction is the same as the preset option, and confirms whether the user operates under the will of the self-determination according to the result of the inquiry content.

A further object of the present invention is to provide a method for authenticating a line of sight in combination with facial authentication, comprising: capturing a user image; extracting a facial feature via the user image, and performing the facial feature and the facial data in the database; Comparing to confirm the user identity; extracting the eye image through the user image, and obtaining the gaze direction of the user from the eye image; recording the gaze direction of the user to obtain a line of sight information; and viewing the line trajectory information and data The trajectory data in the library is compared, and a control signal is sent to an identity verification device when the comparison result is met; and a graphical interface or a hidden interface is provided for the user to gaze at the input, and the graphical interface provides an intelligent inquiry. Answer function, and according to the identity data pre-entered by the user, after generating the corresponding inquiry content, obtain the eye movement or gaze direction of the user, whether the option selected by the user's eye movement or gaze direction is The preset options are the same, and it is confirmed according to the result of the inquiry content whether the user operates under the will.

Further, when recording the gaze direction of the user, a line of sight trajectory input interface is provided for the user to look at the operation to input the line of sight trajectory information according to the gaze direction of the user.

Further, when the line of sight input interface is provided, an auxiliary graphic for the user to look at is provided on the line of sight input interface, so that the user inputs the line of sight information according to the auxiliary figure.

Further, the facial feature is an iris image of an iris region in the eye image of the user.

Further, the facial feature is a facial image of the user The facial feature image in the middle.

Further, the facial feature is a blood vessel distribution image of the scleral region in the eye image of the user.

A further object of the present invention is to provide a computer readable recording medium for storing a line of sight trajectory authentication program. After the program is loaded and executed by a computer, the line trajectory authentication method as described above can be completed.

A further object of the present invention is to provide a computer program product for storing a line of sight trajectory authentication program, which can be completed by the computer and executed to complete the line of sight trajectory authentication method as described above.

The present invention has the following advantageous effects over the prior art:

1. The present invention confirms the user's identity through face recognition and line of sight trajectory, which reduces the possibility that the password is cracked by others. Through the above techniques, it is achieved that the passwords used for network activities and physical activities are not lost, will not be copied, and are guaranteed to be used under free will.

2. The invention can avoid the risk that the third person is cracked because the same group of passwords are reused, and the possibility that the user's password is guessed by the third person can be avoided.

100‧‧‧Sight Track Verification System

1A‧‧‧identity verification equipment

10‧‧‧Image capture device

20‧‧‧ display device

21‧‧‧Display interface

30‧‧‧Control unit

31‧‧‧Face feature recognition module

32‧‧‧Eye Motion Analysis Module

33‧‧‧Graphics generation module

34‧‧‧Sight Track Unlocking Module

35‧‧‧Autonomous Will Confirmation Module

36‧‧‧Database

IN‧‧‧ networked equipment

ME‧‧‧ storage unit

1B‧‧‧ safe

11B‧‧‧ cabinet

12B‧‧‧ protective door

13B‧‧‧door lock module

14B‧‧‧Contact device

141B‧‧‧ window

15B‧‧‧ transmission line

16B‧‧‧ storage box

17B‧‧‧Reel

1C‧‧‧Access Control System

11C‧‧‧ wall mount host

12C‧‧‧door lock module

13C‧‧‧Contact device

14C‧‧‧Security host

15C‧‧‧ gate

16C‧‧‧ transmission line

B1‧‧‧Sight Track Input Interface

B11‧‧‧ cursor

L1‧‧‧Sight trajectory line segment

C1‧‧‧Sight Track Input Interface

C11‧‧‧Guide graphics

L2‧‧‧ Sight line segment

C12‧‧ cursor

D1‧‧‧Sight Track Input Interface

D11‧‧‧ objects

D12‧‧‧ objects

L31‧‧‧ line segment

L32‧‧‧ line segment

L33‧‧‧ line segment

L34‧‧‧ line segment

J‧‧‧ graphical interface

J1‧‧‧ option button

J2‧‧‧ option button

PA‧‧‧Iris alignment graphics

PB‧‧‧Face alignment graphics

PC‧‧‧ scleral alignment

Step S01 to step S11

FIG. 1 is a block diagram of a line of sight trajectory authentication system of the present invention.

Fig. 2 is a schematic view showing a first embodiment of the present invention.

Figure 3 is a schematic view of a second embodiment of the present invention.

Figure 4 is a schematic diagram of the self-willing inquiry menu of the present invention.

FIG. 5 is a schematic diagram (1) of the face feature comparison menu of the present invention.

Figure 6 is a schematic diagram (2) of the face feature comparison menu of the present invention.

Figure 7 is a schematic diagram (3) of the face feature comparison menu of the present invention.

8-1 and 8-2 are schematic diagrams (1) of the line of sight input interface of the present invention.

9-1 and 9-2 are schematic diagrams (2) of the line of sight input interface of the present invention.

10-1 and 10-2 are schematic diagrams (3) of the line of sight input interface of the present invention.

FIG. 11 is a schematic flow chart (1) of a method for authenticating a line of sight of the present invention.

FIG. 12 is a schematic flow chart (2) of the method for authenticating a line of sight of the present invention.

FIG. 13 is a schematic flow chart (3) of the method for authenticating a line of sight of the present invention.

The detailed description and technical contents of the present invention will now be described with reference to the drawings. In addition, the drawings in the present invention are for convenience of description, and the ratios thereof are not necessarily drawn to actual scales, and the drawings and their proportions are not intended to limit the scope of the present invention, and are described herein.

The present invention provides a line-of-sight trajectory authentication system 100 incorporating face authentication, which can be applied to various types of identity verification devices 1A to confirm the identity of the user. The identity verification device 1A, such as a safe, an access control system, an ATM device, a network ATM device, a natural person voucher device, or the like, or other such identity verification device, is not intended to be limiting in the present invention.

The following is a description of a specific embodiment of the line-of-sight track authentication system of the present invention. Please refer to FIG. 1 , which is a block diagram of the line-of-sight track authentication system of the present invention, as shown in the figure:

The gaze track authentication system 100 mainly includes an image capturing device 10, a display device 20, and a control unit 30 connected to the image capturing device 10 and the display device 20.

The image capturing device 10 is configured to capture a user image to further extract facial features and eye images of the user. Specifically, the image capturing device 10 may be a camera equipped with a photosensitive coupled device (CCD) or a complementary metal-oxide semiconductor (CMOS), which is not limited in the present invention. . When the image capturing device 10 captures the user image, the image capturing device 10 transmits the user image to the control unit 30 for analysis.

The display device 20 provides a display interface 21 for the user to look at. The display interface 21 can be a TN panel (Twisted Nematic), a STN panel (Super-twisted nematic display), a VA panel (Vertical Alignment), an IPS panel (In-Plane Switching), and a PLS panel (Plane to Line Switching). , ASV panel (Advanced Super-V), FFS panel (Fringe Field Switching), OCB (Optical Compensated Birefringence), etc. or other such The panel, or the conventional CRT screen (Cathode Ray Tube), is not limited in the present invention.

The control unit 30 can be coupled to the storage unit ME for storing data to perform corresponding steps by accessing a program or data in the storage unit ME. The control unit 30 can be a central processing unit (CPU), or other programmable and general purpose or special purpose microprocessor (Microprocessor), digital signal processor (DSP), Programmable controllers, Application Specific Integrated Circuits (ASICs), Programmable Logic Devices (PLDs), or other similar devices or combinations of these devices. The storage unit ME can be any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash memory (Flash Memory). Or a similar element or a combination of the above elements. The storage unit ME can also be constructed from one or more accessible non-volatile memory components. Specifically, it can be a hard disk, a memory card, or an integrated circuit or a firmware. In another preferred embodiment, the control unit 30 can form a processor together with the storage unit ME. The control unit 30 loads the program pre-stored by the storage unit ME and executes a corresponding algorithm.

For the identity verification device 1A, the following two aspects are described:

In a preferred embodiment, the identity verification device 1A can be a safe 1B. Please refer to FIG. 2 . The safe 1B includes a box 11B, Protective door 12B, door lock module 13B, and eye contact device 14B. The casing 11B has an accommodation space for storing articles. The protective door 12B is movably disposed on one side of the casing 11B to cover the accommodating space, so that the accommodating space is sealed from the outside. The door lock module 13B is used to control the opening or closing of the protection door 12B. The connecting device 14B and the casing 11B are connected by a transmission line 15B. The casing 11B is provided with a storage box 16B and a reel 17B disposed under the storage box 16B. In use, the reel 17B can be wound around the transmission line 15B and stored under the storage case 16B.

The display device 20 and the image capturing device 10 are disposed in the eye-catching device 14B. During operation, the user can place the window 141B on the eye-catching device 14B against the eye of the user, and by looking at the direction. The cursor on the display device 20 is operated to input authentication input information, and the door lock module 13B is unlocked or locked by the result of the comparison.

In another preferred embodiment, the identity verification device 1A can be an access control system 1C. Please refer to FIG. 3 . The access control system 1C includes a wall mount host 11C, a door lock module 12C, and an eye contact device 13C. A security host 14C is installed in the house. The wall mount main body 11C is disposed on one side of the gate 15C or directly on the gate 15C. The door lock module 12C is used to control the opening or closing of the gate 15C. The connection device 13C and the wall-mounted host 11C are connected by a transmission line 16C. The wall-mounted host 11C is provided with a display slot, and a reel disposed on the inner side of the wall-mounted host 11C, when the user does not use it. The cord reel can be wrapped in the wall-mounting host 11C and the eye-catching device 13C can be retracted to the wall-mounted host 11C for the setting slot provided by the eye-catching device 13C.

The display device 20 and the image capturing device 10 are disposed in the eye-catching device 13C. During operation, the user can place the window on the eye-catching device 13C against the eye and operate the display device 20 by looking at the direction. The cursor on the input enters the authentication input information.

The control unit 30 includes a facial feature recognition module 31, an eye movement analysis module 32, a graphic generation module 33, a visual line trajectory unlocking module 34, an autonomous will confirmation module 35, and a database 36.

The database 36 stores user data and a user index, and the user data includes user-defined identity information such as user rights, names, date of birth, or other personal information, and the user index includes The user's track password, iris code, scleral code, facial features code, etc., the user index has a corresponding code for each user, and can perform password identification, password, fingerprint feature, palm feature, or direct The user name input by the I/O device finds the user corresponding to the code, thereby judging the user's authority through the user profile of the user. The database 36 can be pre-set for the user to input user data, and the password or identity information can be input by the I/O device (for example, through a keyboard, a touch screen, etc.) and the image capturing device 10.

The facial feature recognition module 31 is configured to acquire the user image, extract facial features from the user image, and compare the facial features with the facial data in the database 36 to confirm the user identity. . In order to achieve an accurate recognition rate, in a preferred embodiment, the facial features may be a user's iris feature, a facial feature, a scleral feature, or a combination of the above three features (eg, an iris feature ratio) Right, then face features, scleral features, etc.).

The following describes the implementation of iris features, facial features, and scleral features:

In a preferred embodiment, the facial feature recognition module 31 extracts an eye image of the user from the user image to obtain an iris image through the iris region in the eye image. The iris area extracted during facial recognition will be compared with the iris data (face data) in the database 36 to confirm the identity of the user.

The following is a brief description of the method of iris alignment. Since the iris has a high albedo for the frequency band of near-infrared light, in order to facilitate the feature of the iris, the user's eye is illuminated with near-infrared light to capture the iris image. In the continuation, the facial feature recognition module 31 defines an eye region from the facial image of the user, and performs binarization processing on the eye region to obtain an iris to the eye white and the iris to the pupil boundary. Area, which is defined as the iris area. In order to avoid the influence of the change of the iris itself due to the enlargement and reduction of the pupil, the obtained iris can be normalized and then compared. In the continuation, the normalized iris image extracts the iris feature from the filter to represent the iris information, and the extracted information is a two-bit iris code. Finally, the extracted iris code is compared with the iris data in the database 36 to find the most similar iris data to confirm the user's authority.

In addition to the above-described embodiment of performing identity authentication through the iris feature, in another preferred embodiment, the facial feature is the facial image of the user. A facial feature image such as a facial contour, a relative facial position, or a raised portion of the face. The facial feature recognition module 31 can compare the image of the entire face of the user with the facial data in the database 36 to confirm the identity of the user.

Specifically, the facial features can be transmitted through Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Orthogonal Local Preserving Projection (OLPP), Discrete Wavelet Transform, Support Vector Machine, or K-nearest Neighbors can be used to calculate the contour of the face, the relative position of the facial features, or the raised parts of the face. The facial features are compared with the facial data in the database 36 to determine the user's authority.

The technology related to the identification of facial features has been supported by many related technologies in the past, and will not be repeated here.

In another preferred embodiment, the facial feature is a blood vessel distribution image of the scleral region of the user's eye image. In principle, the method of scleral recognition is quite similar to the way of iris recognition, and both regions of interest can be extracted in a binarized manner. The main difference between scleral recognition and iris recognition is that the captured regions are different. The image captured by iris recognition is the iris texture of the human eye, and the image captured by sclera recognition is the blood vessel of the sclera (eye white) in the human eye. distributed.

The following is a brief description of the method of scleral alignment. First, the facial feature recognition module 31 defines an eye region from the facial image of the user, and obtains an eyelid to the sclera by performing binarization (or color separation) on the eye region. The boundary between the sclera and the corneal boundary, defining the region between the two boundaries as the scleral region. The scleral images captured can be normalized for subsequent alignment. In succession, the normalized scleral image extracts the blood vessel distribution through the filter to represent the scleral information, and the extracted information is a two-bit sclera code. Finally, the extracted sclera code is compared with the scleral data in the database 36 to find the most similar scleral data to confirm the user's authority.

After the image capturing device 10 obtains the user image, the captured user image is provided to the eye movement analysis module 32 for analysis to obtain the user's gaze direction and form an operable cursor according to the gaze direction.

When acquiring the user image, the eye movement analysis module 32 obtains the gaze direction and the eye movement of the user according to the eye image of the user image. The gaze direction refers to the coordinate position of the user on the display device 20. The eye movement analysis module 32 can generate eye movement information by analyzing the geometric relationship between the position of the pupil center and the spot to confirm the gaze direction. . Specifically, the eye movement analysis module 32 obtains the pupil center position and the spot position of the user through the binarization process in the captured eye image, and defines the reference line from the pupil center position, and then The pupil center position and the spot generate a connecting line, and the angle between the reference line and the connecting line is measured, and the relative geometric relationship between the center position of the pupil and the spot is determined according to the angle to determine the direction of the gaze. The above process can determine the gaze direction of the user by inserting a plurality of samples of the human eye motion in advance by the trainer, thereby establishing the gaze direction of the user.

The eye movement refers to, for example, the direction of movement of the eyeball, or the eye. The trajectory of the ball, etc., can obtain the eye movement of the user through a simple calculation method, and judge the instruction input by the user. In a preferred embodiment, the eye movement can be obtained through a simple calculation program, for example, dividing the sclera between the two eyes into a plurality of blocks to generate a control signal according to the ratio between the plurality of blocks. . Since related technologies for gaze direction and eye movement detection have been supported by many related cases, they will not be exemplified herein.

The gaze trajectory unlocking module 34 records the gaze direction of the user to obtain a trajectory trajectory information, compares the gaze trajectory information with the trajectory data in the database 36, and generates a control when the comparison result is met. The signal is transmitted to the identity verification device 1A. The eye trajectory unlocking module 34 obtains the gaze direction of the user by the eye movement analysis module 32, and records the coordinates corresponding to each gaze direction, and the code is formed according to the order of the gaze direction and the connection line. The string is compared to the stored trajectory data stored in the database 36 to verify the identity of the user. At startup, the graphics generation module 33 provides a line of sight trajectory input interface for the user to view the operation to input the line of sight trajectory information according to the gaze direction of the user. In a preferred embodiment, the graphics generating module 33 can further provide an auxiliary graphic for the user to look at in the line of sight input interface, so that the user inputs the line of sight information according to the auxiliary graphic.

The autonomous will confirmation module 35 is connected to the graphic generation module 33 to provide a graphical interface or a hidden interface for the user to gaze input, and corresponding to the eye movement analysis module 32 to obtain the user's eye. Department action or gaze direction. The autonomic confirmation module 35 determines the eye movement or the gaze direction of the user. Whether the user operates under his or her will.

In a preferred embodiment, the graphics generation module 33 provides a graphical interface J for the user to use when the autonomous confirmation mode is activated. The graphical interface J provides a smart inquiry function. The inquiry function generates the corresponding inquiry content according to the identity data input by the user in advance, and judges whether the user operates the device according to the result of the inquiry content. Please refer to FIG. 4, the self-determination confirmation module 35 requests a user response according to the user's identity. In the embodiment, the question content is "What is your favorite food?" and provides user correspondence. The option buttons J1 and J2, the option button J1 displays "beef noodles", and the option button J2 displays "Italian noodles". If the user initially sets, the user fills in the favorite food as beef noodles when the user operates. When the option button J2 (Italian face) is selected in the menu, the autonomous will confirm module 35 determines that the user operates the menu in the case of non-autonomous will, and generates a corresponding control signal. In a preferred embodiment, the control signal can be transmitted to, for example, a security device, perform a locking function, and prohibit any access actions. In another preferred embodiment, the control signal can be transmitted to a background center via the network device IN, and the background center is alerted.

In another preferred embodiment, the graphical interface may be a graphic for the user to view and select, so that the user requests the rescue in a hidden state in the case of the non-autonomous will to operate the menu. As shown in "FIG. 9-1" to "FIG. 9-2", the graphic generation module 33 is provided with an emergency call button C13 in the upper right corner of the line-of-sight track input interface C1. When the user operates the device under the condition of being coerced, the user can move to the emergency help button C13 via the gaze direction, and the autonomous will confirm module 35 determines When the user's gaze direction is moved to the emergency call button C13, a warning signal is transmitted to the background center or the cloud rescue center via the network connection device IN.

In another preferred embodiment, the autonomous will confirmation module 35 provides a hidden interface, for example, executed in the background, which is triggered by the user's eye movements. When the user is subjected to external coercion, the corresponding eye action can be input in the menu (for example, according to the gaze direction, regular or irregular eye rotation action, etc.), and the autonomic confirmation module 35 is analyzed through the eye movement. When the module 32 detects that the user performs a predefined eye movement, the autonomous will confirm module 35 generates a corresponding control signal. In a preferred embodiment, the user can determine the rescue command to be transmitted through the eye movement or determine the object to be rescued. For example, the user can contact the police line through specific eye movements and communicate with the police station through the cloud reporting system or the background service system. In the event of a fire, contact the Fire Department through another specific eye movement to request a fire station or back-office system to assist in reporting support. It can also be sent to the back office or cloud rescue system for alert notification.

The following is a description of a specific embodiment. Please refer to FIG. 5, FIG. 6 and FIG. 7 together. In the following, a facial feature comparison menu according to a preferred embodiment of the present invention is disclosed. Please refer to FIG. 5 first:

When the user activates the gaze track authentication system 100, the graphic generation module 33 displays the facial feature verification menu (iris verification menu, facial features verification menu, sclera verification menu) before the display interface 21. In this embodiment, it is used to capture the iris characteristics of the user for comparison. An iris alignment pattern PA is displayed on the iris verification menu. When performing identity verification, the user first closes the eye first. The image capturing device 10 captures the user's eye image by the image capturing device 10, and the captured eye image is displayed on the iris verification menu. At the same time of shooting, the line-of-sight trajectory authentication system 100 activates a near-infrared light-emitting device (not shown) to illuminate the user's eyes. The user can move in front of the image capturing device 10 according to the relative position of the iris alignment pattern PA and the user's eye image to align the iris region in the user's eye image to the middle of the iris alignment pattern PA. The facial feature recognition module 31 extracts the iris image of the user and compares it with the iris data (face data) in the database 36 to complete the identity verification.

For the implementation of the feature comparison, please refer to Figure 6:

In this embodiment, it is used to capture the facial features of the user's face for comparison. When the facial feature recognition is performed, the graphic generation module 33 displays a face alignment graphic PB on the facial features verification menu. The face alignment pattern PB is elliptical and displayed in the middle of the face feature comparison menu. The image capturing device 10 captures a facial image of the user, and the captured facial image is displayed on the facial features verification menu, and the user can determine the relative position of the facial alignment image PB and the user facial image according to the facial alignment image. The image capturing device 10 is moved forward to determine that the contour range of the user can be moved to the inner side of the face alignment pattern PB, so that the facial feature recognition module 31 recognizes the facial features of the user according to the facial features of the user, and the data The facial features (face data) in the library 36 are compared to complete the identity verification.

For the implementation of the scleral feature comparison, please refer to Figure 7:

In this embodiment, it is used to capture the scleral features of the user for comparison. A scleral alignment graphic PC is displayed on the scleral verification menu. Said The scleral alignment pattern PC is rectangular and is placed in the middle of the sclera verification menu. When performing the identity verification, the user first approaches the image capturing device 10 to capture the eye image of the user by the image capturing device 10, and the captured eye image is displayed in the sclera verification. On the menu. The user can move the relative position of the scleral image and the user's eye image in front of the image capturing device 10 to move the entire eye region in the user's eye image to the middle of the scleral alignment graphic PC. The Eli face feature recognition module 31 extracts the scleral image of the user and compares it with the scleral data (face data) in the database 36 to complete the identity verification.

In another preferred embodiment, the facial feature recognition module 31 can directly extract an eye region from the facial image to obtain a scleral image, and perform identity verification by comparing the scleral image.

The following is an explanation of the implementation of the line of sight input interface. Please refer to Figure 8-1 to Figure 8-2:

When the user is authenticated, the gaze track unlocking module 34 displays the gaze track input interface B1 on the display device 20 for the user to look at and input the gaze track according to the moving direction of the line of sight.

The gaze track input interface B1 in this embodiment is a blank window, and the cursor B11 for confirming the gaze direction is displayed on the blank window, and the cursor B11 follows the gaze direction of the user, and determines the trajectory according to the control instruction. The starting point and ending point of the input. When the user inputs a control command, the position where the cursor B11 is located is the starting point of the track input, and the user inputs the control command again. When the cursor B11 is located as the end point of the track and ends the trajectory drawing (as shown in Figure 8-2), a line of sight trajectory line L1 is drawn. The obtained gaze trajectory line segment L1 is compiled into a corresponding set of trajectory codes and stored as trajectory input information, and the trajectory input information is compared with the trajectory data pre-stored by the user in the user index to verify the user identity.

In a preferred embodiment, the control command can be passed through a physical or virtual button for the user to press an input to determine the start and end points of the trajectory. In another preferred embodiment, the control command can determine a starting point and an ending point of the trajectory through a user's eye motion command (eg, blinking).

The following is a description of another preferred embodiment of the line of sight input interface, please refer to "Figure 9-1" to Figure "9-2":

The gaze track input interface C1 of the embodiment provides a plurality of guide patterns C11, and the guide pattern C11 is distributed on the line trajectory input interface C1. The user can align and look at the top when looking at the line trajectory input interface C1. The graphic C11 is directed to guide the user's line of sight by the guiding pattern C11, increasing the accuracy of the line-of-sight trajectory line segment L2 input. When the user's line of sight (cursor C12) moves to the guide pattern C11, the guide pattern C11 changes the graphical feature (for example, anti-black) for the user to confirm the position of the gaze direction movement, so that the user confirms the input trajectory and completes the input. A line of sight line segment L2 is obtained, and the line of sight line segment L2 is compiled into a corresponding set of track codes, stored as track input information, and compared with the track data pre-stored by the corresponding user to verify the user identity.

The following is another preferred embodiment for the line of sight trajectory input interface. For instructions, please refer to "Figure 10-1" to "Figure 10-2":

The line-of-sight trajectory input interface D1 of the embodiment provides a plurality of objects for clicking, for the user to move the cursor D13 and watch the point selection. When the user's gaze direction stays on the object D11, a timer of the system is started, and the dwell time of the user's line of sight is calculated. When the dwell time exceeds the set value, the object D11 selected by the user is confirmed, and the gaze direction of the user is Can be moved to the next object D12. When the user's line of sight moves to the next object D12 and the selection is completed, the next object D12 and the previous object D11 are actively connected into a line segment L31. Specifically, as shown in Figure 10-2, the user's gaze direction is in Jupiter (starting point) - Saturn - Uranus - Sun - Earth (end point), when the user's line of sight moves from Jupiter to Saturn and completes the selection. When entering the line segment L31; when the user's line of sight moves from Saturn to Uranus and completes the selection, enter a line segment L32; when the user's line of sight moves from Uranus to the sun and completes the selection, enter a line segment L33; When the line of sight moves from the sun to the earth and completes the selection, it enters a line segment L34. When all the line segments are input, the line of sight unlocking module 34 compiles all the line segments L31-L34 into a corresponding set of track codes and stores them as track input information, and inputs the track input information with the corresponding user pre-stored track. The data is compared to verify the identity of the user.

The following is a description of the gaze track authentication method of the present invention. Please refer to FIG. 11 and FIG. 12 together, which is a schematic flowchart (1) and a flow chart (2) of the gaze track authentication method of the present invention. Show:

The gaze trajectory authentication method can be used in conjunction with the hardware use of the gaze trajectory authentication system 100, and the gaze trajectory authentication method includes the following steps:

After the gaze track authentication system 100 is activated, the gaze track authentication system 100 activates the image capturing device 10 to capture a user image (step S01).

In the continuation, the gaze track authentication system 100 extracts the face feature via the user image, compares the face feature with the face data in the database 36 to confirm the user identity (step S02). When confirming that the user's identity has authority, proceed to the next step to perform the graphic authentication process. When it is determined that the user identity does not have the authority, an error authentication message is generated (step S03). The facial feature may be an iris image of an iris region in the user's eye image, or a facial contour, a facial feature relative position, a facial convex portion, or the user's eye in the user's facial image. An image of blood vessels in the scleral region of the image. In another preferred embodiment, the user image authentication may also be set after the password input is completed, which is not limited in the present invention.

When it is confirmed in the previous step that the user has the authority, the graphic generation module 33 provides a line-of-sight track input interface for the user to look at the operation (step S04). In a preferred embodiment, the line of sight input interface can provide an auxiliary graphic for the user to view, so that the user inputs the line of sight information according to the auxiliary graphic.

When the gaze track input interface is provided on the display device 20, the eye movement analysis module 32 extracts the eye image through the user image, and obtains the gaze direction of the user from the eye image (step S05). At this time, the graphic generation module 33 provides a cursor to follow the gaze direction of the user to operate the line of sight trajectory input interface by the gaze direction of the user.

In the continuation, the line of sight unlocking module 34 records the gaze of the user. The direction is to obtain a line of sight trajectory information (step S06).

Finally, the line-of-sight track unlocking module 34 compares the line-of-sight track information with the track data in the database 36 to confirm whether the line-of-sight track information conforms to the track data in the database 36 (step S07). When the match is confirmed, a control signal is supplied to the identity verification device 1A to start the corresponding program (step S08); if the discrepancy is confirmed, an error authentication message is generated (step S09).

During the execution of the above process, or before or after verification, the autonomous will confirmation module 35 can provide a graphical interface or a hidden interface for the user to gaze input to the eye of the user. The action or gaze direction confirms whether the user is operating under his or her will. Please refer to FIG. 13 together. In a preferred embodiment, when the autonomous will confirmation process is executed, the hidden interface is executed in the background, and the eye movement or gaze direction and the trajectory in the database 36 are performed. The data is compared (step S10), and a control signal is generated when the user's eye movement conforms to the predefined condition (step S11).

In another preferred embodiment, the autonomous will confirmation module 35 can randomly enter the random number, and the query asks the user to respond to the problem on the graphical interface, by the user's eye for the question and answer question. The action of the department confirms whether the user operates the menu under his or her will, and operates the menu when the user is determined to be involuntary will, and generates a corresponding control signal.

The method steps described above can be implemented by a computer readable recording medium, such as a read-only memory, a flash memory, a floppy disk, a hard disk, a compact disk, a flash drive, a magnetic tape. Information accessible by the network Libraries or those skilled in the art can easily think of storage media having the same function. When the computer loads the line of sight trace authentication program and executes it, the method described above can be completed.

In addition to the computer readable recording medium, the above method steps can also be implemented as a computer program product for storing on a network server's hard disk, memory device, such as app store, google play, windows market, or other A similar online application publishing platform can be implemented by uploading a computer program product to a server for downloading by a user.

In summary, the present invention confirms the user's identity through face recognition and line of sight trajectory, which reduces the possibility that the password is cracked by others. Through the above techniques, it is achieved that the passwords used for network activities and physical activities are not lost, will not be copied, and are guaranteed to be used under free will. The invention can avoid the risk of the third person being cracked due to the repeated use of the same set of passwords, and can also reduce the possibility that the user's password is guessed by the third party.

The invention has been described in detail above, but the foregoing is only a preferred embodiment of the invention, and is not intended to limit the scope of the invention, Variations and modifications are still within the scope of the patents of the present invention.

100‧‧‧Sight Track Verification System

1A‧‧‧identity verification equipment

10‧‧‧Image capture device

20‧‧‧ display device

30‧‧‧Control unit

31‧‧‧Face feature recognition module

32‧‧‧Eye Motion Analysis Module

33‧‧‧Graphics generation module

34‧‧‧Sight Track Unlocking Module

35‧‧‧Autonomous Will Confirmation Module

36‧‧‧Database

IN‧‧‧ networked equipment

ME‧‧‧ storage unit

Claims (16)

A sight line trajectory authentication system, comprising: an image capturing device for capturing user images; a display device providing a display interface for the user to look at; and a control unit comprising: an eye movement analysis module Obtaining a gaze direction of the user according to the eye image of the user image; a line of sight trajectory unlocking module records the gaze direction of the user to obtain a trajectory trajectory information, and the trajectory trajectory information and the trajectory data in the database are performed. Comparing and generating a control signal to an identity verification device when the comparison result is met; and an autonomous confirmation module providing a graphical interface or a hidden interface for the user to gaze at the input, the graphical interface providing a wisdom The inquiry function, and according to the identity data input by the user in advance, after generating the corresponding inquiry content, obtaining the eye movement or the gaze direction of the user, whether the option selected by the user's eye movement or gaze direction is As with the pre-set options, depending on the results of the inquiry, the user is determined whether to enter the autonomy Operation. A line-of-sight trajectory authentication system combining face authentication includes: an image capturing device for capturing a user image; and a display device providing a display interface for the user to look at; a control unit, comprising: a facial feature recognition module, the user image is obtained, and the facial features are extracted from the user image, and the facial features are compared with the facial data in the database. To verify the identity of the user; an eye movement analysis module obtains the gaze direction of the user according to the eye image of the user image; and a line of sight trajectory unlocking module records the gaze direction of the user to obtain a line of sight trajectory, the line of sight The trajectory information is compared with the trajectory data in the database, and a control signal is sent to an identity verification device when the comparison result is met; and an independent confirmation module provides a graphical interface or a hidden interface for the user. Looking at the input, the graphical interface provides a smart inquiry function, and according to the identity data previously input by the user, after generating the corresponding inquiry content, obtaining the eye movement or gaze direction of the user, by using the user's eye Whether the selected option of the action or the gaze direction is the same as the preset option, and determines whether the user is based on the result of the inquiry content. Operate under independent will. The gaze track authentication system according to any one of claims 1 to 2, wherein the control unit comprises a graphic generation module, wherein the graphic generation module provides a line of sight trajectory input interface for the user to look at the operation. The sight trajectory information is input according to the gaze direction of the user. The gaze track authentication system of claim 3, wherein the graphics generation module provides an auxiliary graphic for the user to look at in the line of sight input interface, so that the user inputs the line trajectory information according to the auxiliary graphic. The gaze track authentication system according to any one of claims 1 to 2, wherein the facial feature is an iris image of an iris region in an eye image of the user. The gaze track authentication system according to any one of claims 1 to 2, wherein the facial feature is a facial feature image in a facial image of the user. The gaze track authentication system according to any one of claims 1 to 2, wherein the facial feature is a blood vessel distribution image of a scleral region in the eye image of the user. A method for trajectory trajectory authentication includes: capturing a user image; extracting an eye image through the user image, and obtaining a gaze direction of the user from the eye image; recording a gaze direction of the user to obtain a line of sight trajectory information; Comparing the trajectory trajectory information with the trajectory data in the database, and generating a control signal to the identity verification device when the comparison result is met; and providing a graphical interface or a hidden interface for the user to gaze input, The graphical interface provides a smart inquiry function, and according to the identity data previously input by the user, after generating the corresponding inquiry content, obtaining the eye movement or gaze direction of the user, by the user's eye movement or gaze The option selected by the direction is the same as the preset option, and it is confirmed according to the result of the inquiry content whether the user operates under the will. A method for authenticating a line of sight in combination with facial authentication includes: capturing a user image; extracting a facial feature via the user image, comparing the facial feature with facial data in the database to confirm the user identity; The user image extracts the eye image, and obtains the gaze direction of the user from the eye image; records the gaze direction of the user to obtain a line of sight trajectory information; and compares the gaze track information with the trajectory data in the database, And generating a control signal to the identity verification device when the comparison result is met; and providing a graphical interface or a hidden interface for the user to gaze at the input, the graphical interface provides a smart inquiry function, and is pre-entered according to the user The identity data, after generating the corresponding inquiry content, obtain the eye movement or gaze direction of the user, and whether the option selected by the user's eye movement or gaze direction is The options set first are the same, and it is confirmed according to the result of the inquiry content whether the user operates under the will. The gaze trajectory authentication method according to any one of claims 8 to 9, wherein when the gaze direction of the user is recorded, a trajectory trajectory input interface is provided for the user to look at the operation to input the gaze according to the gaze direction of the user. Sight track information. The gaze trajectory authentication method according to claim 10, when the trajectory trajectory input interface is provided, an auxiliary graphic for the user to gaze is provided on the trajectory trajectory input interface, so that the user inputs the gaze trajectory according to the auxiliary trajectory. News. The gaze track authentication method according to any one of claims 8 to 9, wherein the facial feature is an iris image of an iris region in an eye image of the user. The gaze track authentication method according to any one of claims 8 to 9, wherein the facial feature is a facial feature image in a facial image of the user. The gaze track authentication method according to any one of claims 8 to 9, wherein the facial feature is blood of a scleral region in an eye image of the user. Tube distribution image. A computer-readable recording medium for storing a line-of-sight track authentication program, which is loaded into the program by a computer and executed, and the line-of-sight track authentication method according to any one of claims 8 to 14 can be completed. A computer program product for storing a line of sight trajectory authentication program, which is loaded into the program by a computer and executed, and the line of sight trajectory authentication method according to any one of claims 8 to 14 can be completed.