KR102243341B1 - Method for authenticating user using touch pattern - Google Patents

Abstract

Disclosed is a user authentication method using an input touch pattern. The user authentication method for authenticating a user using a calculated similarity comprises the steps of: receiving sensing data for a touch pattern inputted on a screen of a user terminal; grouping the received plurality of pieces of sensing data; calculating a representative value for each parameter of the sensing data with respect to the plurality of pieces of sensing data included in each of the groups; generating a touch pattern profile for each group using the extracted representative value for each parameter; calculating a similarity between the generated touch pattern profile and a previously stored customer registration pattern profile received from the user terminal in the past; and determining whether to authenticate a service requested by the user terminal based on the calculated similarity.

Description

User authentication method using touch pattern {Method for authenticating user using touch pattern}

The present invention relates to a user authentication method using a touch pattern. Specifically, the present invention relates to a method of grouping sensing data for a touch pattern input to a user terminal and determining whether to authenticate a service by calculating a similarity between the touch pattern profile of each group and the customer registration pattern profile stored in advance.

The content described in this section merely provides background information on the present embodiment and does not constitute the prior art.

Recently, due to the development of smart devices and networks, and the development of various network services, various tasks, including banking, which were conventionally performed face-to-face, have been converted to non-face-to-face business processing using online/wireless. In this case, when user authentication is required to perform non-face-to-face business, a user authentication method in which an authentication password input through a user terminal is compared with an authentication password of a user registered in advance is widely used.

The user authentication method can be performed using various authentication means. For example, user authentication methods include a method of using a password based on letters or numbers, a method of using a touch pattern, a method of using a biometric authentication method, and the like.

Among them, the conventional user authentication method using a password (eg, PIN) based on a letter or number has been widely used because it is inexpensive and convenient, but there is an inconvenience of having to memorize the password all the time. In addition, when passwords are exposed due to user carelessness or hacking, financial transactions are possible by others, thereby increasing the risk of financial fraud.

On the other hand, the user authentication method using a touch pattern has an advantage that it is easier to remember than a password based on letters or numbers because a graphical password is used. However, the number of cases is smaller than that of a password based on letters or numbers, and there is a limit to the type of touch pattern actually used. In addition, since it is possible to infer a touch pattern (Smudge attack) by another person's skin surfing or drag traces remaining on the terminal screen, there is a problem that it is somewhat vulnerable to security.

Therefore, there is a need for a user authentication method that can maintain user convenience and enhance the security of pattern authentication.

An object of the present invention is to group sensing data of the touch pattern by deriving a line connecting two adjacent points from a touch pattern input by a user, and a touch pattern profile generated based on the grouped sensing data and registered in advance. It provides a method of authenticating a user by calculating the degree of similarity between the customer registration pattern profiles.

In addition, another object of the present invention is to group sensing data extracted from a touch pattern input by a user using a reference time unit or a reference number of preset events, and a touch pattern profile generated based on the grouped sensing data It is to provide a method of authenticating a user by calculating the similarity between the customer registration pattern profile registered in advance.

In addition, another object of the present invention is a method of calculating the similarity between the touch pattern profile calculated by the above-described method and the customer registration pattern profile using a pre-learned deep learning module, and authenticating a user using the calculated similarity. Is to provide.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention that are not mentioned can be understood by the following description, and will be more clearly understood by examples of the present invention. In addition, it will be easily understood that the objects and advantages of the present invention can be realized by the means shown in the claims and combinations thereof.

An aspect of a user authentication method according to an embodiment of the present invention includes receiving sensing data for a touch pattern input on a screen of a user terminal, grouping the received plurality of sensing data, and each grouped For a plurality of sensing data included in a group, calculating a representative value for each parameter of the sensing data, generating a touch pattern profile for each group using the extracted representative value for each parameter, the Comprising the step of calculating the similarity between the generated touch pattern profile and the previously stored customer registration pattern profile received from the user terminal in the past, and determining whether to authenticate the service requested by the user terminal based on the calculated similarity do.

In addition, the touch pattern is formed so as to pass through a plurality of areas respectively set around a plurality of preset points, and the grouping step includes a plurality of sensing derived from a line connecting two adjacent points within the touch pattern. It may include setting the data as a group.

In addition, the touch pattern is formed so as to pass through the first to third areas respectively set around the first to third points set in advance, and the grouping may include the first area set around the first point Sensing between the first starting point and the first ending point by using the first sensing data of for as a first starting point and the second sensing data for the second area set around the second point as a second ending point Data is set as a first group, the second sensing data is used as a second starting point, and third sensing data for the third area set around the third point as a second ending point, the first starting point And setting sensing data between the and the second end point as a second group.

In addition, the sensing data may be generated in a preset time unit, and some of the sensing data included in the first group and the second group may overlap with each other.

In addition, the number of groups grouped in the touch pattern may be smaller than the number of regions through which the touch pattern passes.

In addition, the step of determining whether to authenticate may include approving the authentication of the service when the similarity is higher than a preset reference value and the touch pattern matches a preset registration pattern.

In addition, the touch pattern includes a plurality of sensing data arranged in the order of generation time, and the grouping comprises generating a group using sensing data included in a preset reference time unit, wherein each group is , Based on different starting points, a range may be set using the reference time unit.

In addition, the touch pattern includes a plurality of sensing data arranged in the order of generation time, and each of the sensing data is divided into events of a predetermined type, and the grouping includes a predetermined reference number of the events. A group is created to include sensing data corresponding to, and each group may have a range set based on a different starting point using the reference number of events.

In addition, the events are classified into any one type of onfling, scrolling, and single tapping, and each of the events can be classified by a finger touch operation (onDown).

In addition, the step of calculating the representative value for each parameter includes extracting the event of the same type within the same group, and calculating the representative value for each parameter with respect to the sensing data included in the extracted event of the same type. Including calculating, in the step of generating a touch pattern profile for each group, the touch pattern profile may include one vector generated by connecting each event type composed of the representative value.

In addition, the parameter of the sensing data is composed of at least two or more of a touch time, a touch position, a touch pressure, a touch size, data of a gyro sensor, or data of an acceleration sensor, and calculating the representative value comprises: It may include calculating an average value, a maximum value, a minimum value, a standard deviation, a median value, a root mean square (RMS), or a root sum of squares (RSS) of a parameter.

In addition, the calculating of the similarity may include calculating the similarity of each group for the group-specific touch pattern profile derived from the touch pattern and the corresponding pre-stored customer registration pattern profile. Including calculating an average, and determining whether or not to authenticate, when the calculated average is higher than a preset reference value, it is determined that authentication has been successful, and when the calculated average is less than a preset reference value, additional authentication May include requesting.

In addition, the step of calculating the similarity includes calculating the similarity for each group using a pre-learned deep learning module, wherein the deep learning module includes the touch pattern profile and the customer registration pattern profile as an input node. A weight of nodes and edges between the input node and the output node, including an input layer denoted by, an output layer having the similarity as an output node, and at least one hidden layer disposed between the input layer and the output layer. May be updated by the learning process of the deep learning module.

On the other hand, another aspect of the user authentication method according to an embodiment of the present invention, the step of receiving sensing data for a touch pattern input on the screen of the user terminal, the step of grouping the received plurality of sensing data, the Generating a touch pattern profile for each group based on a plurality of sensing data included in each grouped group, similarity between the generated touch pattern profile and a customer registration pattern profile previously stored and previously stored from the user terminal And determining whether to authenticate the service requested by the user terminal based on the calculated similarity, wherein the grouping step is derived from a line connecting two adjacent points in the touch pattern. And setting sensing data corresponding to a preset reference time unit or a preset reference number of events as one group.

In the user authentication method of the present invention, sensing data of the touch pattern is grouped by deriving a line connecting two adjacent points from a touch pattern input by a user, and a touch pattern profile generated based on the grouped sensing data By calculating and comparing the degree of similarity between registered customer registration pattern profiles, it is possible to enhance the security of authentication compared to the method of comparing only the matching of the existing touch patterns. At the same time, since the same input interface as the method using the existing touch pattern is used, the inconvenience that the user has to adapt to the new authentication environment can be eliminated.

In addition, in the user authentication method of the present invention, the sensing data extracted from the touch pattern input by the user is grouped using a reference time unit or a reference number of preset events, and a touch pattern generated based on the grouped sensing data By calculating and comparing the degree of similarity between the profile and the previously registered customer registration pattern profile, it is possible to improve the security of authentication. In addition, since a separate pattern input step for authentication can be omitted, user convenience can be improved, and accuracy of authentication can be improved by analyzing the similarity using a touch pattern that the user normally inputs.

In addition, the user authentication method of the present invention calculates the similarity between the touch pattern profile calculated by the above-described method and the customer registration pattern profile using a pre-learned deep learning module, and authenticates the user using the calculated similarity, The accuracy and security of user authentication can be improved.

In addition to the above-described contents, specific effects of the present invention will be described together while describing specific details for carrying out the present invention.

1 is a conceptual diagram illustrating a system for performing a user authentication method according to an embodiment of the present invention.
2 is a flowchart illustrating a user authentication method according to some embodiments of the present invention.
3 is a flowchart illustrating a user authentication method according to an embodiment of the present invention.
4 and 5 are diagrams for explaining an example of a user authentication method according to an embodiment of the present invention.
6 is a diagram illustrating an example of the present invention for calculating a similarity in step S260 of FIG. 3.
FIG. 7 is a diagram illustrating another example of the present invention in which a similarity degree is calculated using a deep learning module in step S260 of FIG. 3.
8 is a flowchart illustrating a user authentication method according to another embodiment of the present invention.
FIG. 9 is a diagram illustrating an example of the present invention in which sensing data is grouped using a reference time unit in step S330 of FIG. 8.
10 and 11 are diagrams for explaining another example of the present invention in which sensing data is grouped using a gesture type in step S330 of FIG. 8.
FIG. 12 is a diagram illustrating a method of calculating a similarity by using a deep learning module in step S350 of FIG. 8.
13 is a diagram showing the configuration of a deep learning module used in some embodiments of the present invention.
14 is a diagram illustrating a hardware implementation of a system that performs a user authentication method according to some embodiments of the present invention.

The terms or words used in this specification and claims are limited to their general or dictionary meanings and should not be interpreted. According to the principle that the inventor can define the concept of terms or words in order to explain his own invention in the best way, it should be interpreted as a meaning and concept consistent with the technical idea of the present invention. In addition, the embodiments described in the present specification and the configurations shown in the drawings are only one embodiment in which the present invention is realized, and do not represent all the technical ideas of the present invention, and thus they can be replaced at the time of application. It is to be understood that there may be various equivalents and variations and examples of application.

Terms such as first, second, A, and B used in the specification and claims may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another component. For example, without departing from the scope of the present invention, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element. The term'and/or' includes a combination of a plurality of related listed items or any of a plurality of related listed items.

The terms used in the specification and claims are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In the present application, terms such as "comprise" or "have" should be understood as not precluding the possibility of existence or addition of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification. .

Unless otherwise defined, all terms including technical or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs.

Terms as defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and should not be interpreted as an ideal or excessively formal meaning unless explicitly defined in the present application. Does not.

In addition, each configuration, process, process, or method included in each embodiment of the present invention may be technically shared within a range that does not contradict each other.

Hereinafter, a user authentication method according to an embodiment of the present invention and a system for performing the same will be described in detail with reference to FIGS. 1 to 14.

1 is a conceptual diagram illustrating a system for performing a user authentication method according to an embodiment of the present invention.

Referring to FIG. 1, a system according to an embodiment of the present invention includes an authentication server 100 and a user terminal 200.

The authentication server 100 (hereinafter, the server) provides financial services to the user through the user terminal 200, and checks the identity of the user by using the touch pattern received from the user terminal 200 before providing the financial service. User authentication can be performed.

At this time, the server 100 generates a touch pattern profile using the input touch pattern, and calculates the similarity between the generated touch pattern profile and the previously registered touch pattern profile (hereinafter, customer registration pattern profile) to confirm the user's identity. Alternatively, you can perform self-authentication.

However, the user authentication method performed in the server 100 is not limited to the above operation, and it is obvious that it can be applied and performed in various embodiments, but hereinafter, for convenience of explanation, the user's identity authentication in financial transactions It will be described using an example of performing.

The server 100 may operate as an execution subject of the user authentication method. Specifically, the server 100 may receive sensing data for a touch pattern input on the screen of the user terminal 200.

Here, the sensing data represents touch data collected when the user touches the user terminal 200. For example, the sensing data includes the time, position, pressure, and size of a touch input by the user on the touch panel of the user terminal 200. In addition, the sensing data may include data sensed by a gyro sensor (not shown) or an acceleration sensor (not shown) provided in the user terminal 200 while the user inputs a touch.

Subsequently, the server 100 may group sensing data of the received touch pattern.

In an embodiment of the present invention, when a touch pattern is received through a pattern authentication request pop-up, the server 100 may group sensing data based on a preset area.

Specifically, the server 100 may receive a user's touch pattern through a pattern authentication request pop-up composed of a plurality of preset points. In this case, the touch pattern may be formed so as to pass through a plurality of regions respectively set around a plurality of preset points. The server 100 may derive sensing data for a plurality of points and a touch pattern formed in a plurality of areas.

Subsequently, the server 100 may set a plurality of sensing data derived from a line connecting two adjacent points in the touch pattern as one group. For example, in the case of a touch pattern connecting a first point and a second point, the server 100 senses data on a line formed between the first area including the first point and the second area including the second point. Can be derived. Subsequently, the server 100 may set the derived sensing data as one group.

In this case, the number of groups set by the server 100 may be smaller than the number of areas through which the touch pattern passes. A process of grouping using a preset area will be described in detail below with reference to FIGS. 3 to 6.

In addition, in another embodiment of the present invention, when sensing data for a touch pattern generated in an application installed in the user terminal 200 is received, the server 100 determines a reference number of events in a reference time unit or a predetermined type. Sensing data can be grouped by using.

For example, when the reference time unit is 60 seconds, the server 100 may set all touch patterns occurring within 60 seconds within the application as one group. In this case, a range of the generated groups may be set using a reference time unit based on different starting points. The process of grouping using the reference time unit will be described later with reference to FIG. 9.

As another example, the server 100 may group sensing data using a reference number of events of a predetermined type. The server 100 may classify the sensing data received for the touch pattern into a predetermined type of event. Here, the event means a type of touch, and may include types of onfling, onscroll, and single tap (onsingletap). However, the present invention is not limited thereto, and of course, various types of events not described in the present specification may be added.

Subsequently, the server 100 may generate a group including sensing data corresponding to the reference number of predetermined events. In this case, a range of the generated group may be set based on different starting points, using a reference number of events. The process of grouping using events will be described later with reference to FIGS. 10 and 11.

Subsequently, the server 100 calculates a representative value for each parameter of the sensing data for a plurality of sensing data included in each group grouped through each of the above-described methods, and uses the calculated representative value to calculate a touch pattern for each group. You can create a profile.

Subsequently, the server 100 may calculate a similarity between the generated touch pattern profile and the customer registration pattern profile previously received and stored from the user terminal 200.

For example, the server 100 may calculate a group-by-group similarity to a touch pattern profile for each group and a customer registration pattern profile corresponding thereto. In this case, the algorithm for calculating the similarity may be variously modified and used, and a detailed description of the algorithm is widely known to those of ordinary skill in the art, and a detailed description thereof will be omitted herein. Subsequently, the server 100 may calculate an average similarity for the calculated similarity for each group.

As another example, the server 100 may calculate the similarity by using a deep learning module that has been learned in advance. In this case, the pre-learned deep learning module may receive a touch pattern profile and a customer registration pattern profile, and output a similarity.

Subsequently, the server 100 may determine whether to authenticate the service based on the calculated similarity.

In the present invention, the server 100 and the user terminal 200 may be implemented as a server-client system. Specifically, the server 100 may classify, store, and manage a touch pattern and a pre-generated customer registration pattern profile for each user account, and provide various services related to financial services and financial information provision to the user terminal 200. It can be provided through the installed terminal application.

In this case, the terminal application may be a dedicated application for receiving a touch pattern or a web browsing application. Here, the dedicated application may be an application embedded in the user terminal 200 or an application downloaded from an application distribution server and installed in the user terminal 200.

The user terminal 200 refers to a communication terminal capable of operating an application in a wired or wireless communication environment. In FIG. 1, the user terminal 200 is shown as a smart phone, which is a kind of portable terminal, but the present invention is not limited thereto, and may be applied without limitation to a device capable of operating a financial application as described above. have. For example, the user terminal 200 may include various types of electronic devices such as a personal computer (PC), a notebook computer, a tablet, a mobile phone, a smart phone, and a wearable device (eg, a watch-type terminal).

In addition, although only one user terminal 200 is shown in the drawings, the present invention is not limited thereto, and the server 100 may operate in conjunction with a plurality of user terminals 200.

Additionally, the user terminal 200 includes an input unit for receiving a user's input, a display unit for displaying visual information, a communication unit for transmitting and receiving signals to and from the outside, a camera unit for photographing a user's face, and the user's voice as digital data. It may include a microphone unit that converts, and a control unit that processes data, controls each unit inside the user terminal 200, and controls data transmission/reception between the units. Hereinafter, commands executed by the control unit inside the user terminal 200 according to a user's command are collectively referred to as being executed by the user terminal 200.

Meanwhile, the communication network 300 serves to connect the server 100 and the user terminal 200. That is, the communication network 300 refers to a communication network that provides an access path so that the user terminal 200 can transmit and receive data after accessing the server 100. The communication network 300 is, for example, wired networks such as LANs (Local Area Networks), WANs (Wide Area Networks), MANs (Metropolitan Area Networks), ISDNs (Integrated Service Digital Networks), wireless LANs, CDMA, Bluetooth, satellite communication, etc. It may cover wireless networks of, but the scope of the present invention is not limited thereto.

Hereinafter, a user authentication method performed in a system according to an embodiment of the present invention will be described in detail.

2 is a flowchart illustrating a user authentication method according to some embodiments of the present invention.

Referring to FIG. 2, the server 100 receives sensing data for a touch pattern input on the screen through the user terminal 200 (S110). Here, the sensing data may include a user's touch time, a touch position, a touch pressure, a touch size, and data of a gyro sensor or an acceleration sensor provided in the user terminal 200.

Subsequently, the server 100 groups the received sensing data (S120). In this case, the server 100 may group the sensing data using a preset area, a preset reference time unit, or a preset type of event.

For example, when receiving sensing data for a touch pattern connecting a plurality of preset points, the server 100 collects a plurality of sensing data derived from a line connecting two adjacent points in the touch pattern into one group. Can be set.

As another example, the server 100 may apply a preset reference time unit to sensing data arranged in the order of generation time. The server 100 may set sensing data included in a preset reference time unit as a group.

As another example, the server 100 may classify sensing data arranged in the order of generation time into predetermined types of events (eg, throwing, scrolling, single tap, etc.). The server 100 may group sensing data corresponding to the reference number of preset events into one group.

Subsequently, the server 100 calculates a representative value for each parameter of the sensing data for each group (S130). Here, the parameter means each data on data of a touch time, a touch position, a touch pressure, a touch size, a gyro sensor or an acceleration sensor included in the above-described sensing data. The server 100 may calculate a representative value by calculating an average value, a maximum value, a minimum value, a standard deviation, a median value, a root mean square (RMS) or a root sum square (RSS) of each parameter. However, the method of calculating the representative value is only an example of the present invention, and the present invention is not limited thereto.

Subsequently, the server 100 generates a touch pattern profile using the calculated representative value (S140).

Subsequently, the server 100 calculates a similarity between the generated touch pattern profile and the previously stored customer registration pattern profile (S150).

For example, the server 100 may calculate a similarity for each group and an average of the calculated similarity for a group-specific touch pattern profile derived from a touch pattern and a corresponding pre-stored customer registration pattern profile. .

As another example, the server 100 may calculate a group-specific similarity for a group-specific touch pattern profile and a customer registration pattern profile using a pre-learned deep learning module, and calculate an average of the calculated similarity.

Subsequently, the server 100 determines whether to authenticate the service based on the calculated similarity (S160).

In this case, when the calculated similarity is greater than a preset reference value, the server 100 may determine that authentication is successful and provide a service to the user terminal 200.

On the other hand, when the calculated similarity is less than a preset reference value, the server 100 may provide a pop-up requesting additional authentication to the user terminal 200.

Hereinafter, a user authentication method performed by grouping sensing data using a preset area according to an embodiment of the present invention will be described in detail.

3 is a flowchart illustrating a user authentication method according to an embodiment of the present invention. Hereinafter, it is assumed that the touch pattern is formed so as to pass through a plurality of regions respectively set around a plurality of preset points.

Referring to FIG. 3, the server 100 receives a touch pattern initially input by a user through the terminal screen SC of the user terminal 200 (S210). In this case, the server 100 may receive the first registered touch pattern through the initial touch pattern registration pop-up displayed on the user terminal 200.

Subsequently, the server 100 extracts a touch pattern profile for the registered touch pattern (S220).

Specifically, the server 100 may group by deriving sensing data for a plurality of points and a plurality of areas on which a registered touch pattern is formed. Subsequently, the server 100 may calculate a representative value for each parameter of sensing data for the derived group, and extract a registration touch pattern profile based on the calculated representative value.

Subsequently, the server 100 generates a customer registration pattern profile based on the extracted touch pattern profile (S230). In this case, the server 100 may store the user's account information, the registered touch pattern registered by the user, and the customer registration pattern profile for the registration touch pattern as one.

Subsequently, when receiving a financial transaction request from the user terminal 200, the server 100 requests the user to authenticate the touch pattern (S240). In this case, the server 100 may display a pattern authentication pop-up on the terminal screen SC.

Subsequently, the server 100 extracts a touch pattern profile based on the touch pattern input through the touch pattern authentication pop-up (S250).

Specifically, the server 100 may derive and group sensing data for a plurality of points and a plurality of regions on which an input touch pattern is formed, and set sensing data of a line connecting two adjacent points as one group. The server 100 may calculate a representative value for each parameter of the derived group sensing data, and extract an input touch pattern profile based on the calculated representative value.

Subsequently, the server 100 calculates whether or not the received touch pattern matches the registered touch pattern, and a similarity between the extracted touch pattern profile and the customer registration pattern profile (hereinafter, touch pattern similarity) (S260).

In this case, the server 100 may compare the registered touch pattern of the customer registration pattern profile with the touch pattern input through the pattern authentication pop-up to check whether they match. In addition, the server 100 may calculate a group-by-group similarity of the input touch pattern and a corresponding customer registration pattern profile, and calculate an average value of the calculated similarity by group.

However, the present invention is not limited thereto, and the touch pattern similarity may be calculated by a pre-learned deep learning module.

Subsequently, the server 100 determines whether the input touch pattern and the registered touch pattern match (S270), and if the touch patterns do not match, the server 100 may return to step S240 and request re-authentication for the touch pattern. .

On the other hand, when the touch patterns match, the server 100 determines whether the similarity of the touch pattern is greater than the reference value (S280).

Subsequently, when the similarity of the touch pattern is greater than the preset reference value, the server 100 may execute a financial transaction requested through the user terminal 200 (S291).

On the other hand, when the similarity of the touch pattern is less than the preset reference value, the server 100 may request additional authentication using the user terminal 200 (S293).

Hereinafter, a method for grouping sensing data using a preset area according to an embodiment of the present invention will be described in detail.

4 and 5 are diagrams for explaining an example of a user authentication method according to an embodiment of the present invention. Here, FIG. 4 is a diagram illustrating sensing data of a touch pattern, and FIG. 5 is a diagram illustrating a method of extracting a touch pattern profile using an input touch pattern.

Referring to FIG. 4, the server 100 may receive a touch pattern TP through the terminal screen SC of the user terminal 200.

Specifically, the server 100 may receive the shape of the touch pattern TP and sensing data for the touch pattern TP. The server 100 may set a plurality of areas for a plurality of preset points, respectively.

For example, the server 100 may set the point area ar3 with respect to the area around the specific point d3 around the specific point d3. Subsequently, when a part of the touch pattern TP is included in the point area ar3, the server 100 may use sensing data for the point area.

At this time, the sensing data includes touch coordinates, touch pressure, touch area, touch time, sliding information (speed, distance, time and angle) for the touch pattern TP included in the point area ar3, gyro sensor data, and acceleration sensor. May contain data.

More specifically, <a1> of FIG. 5 is a diagram illustrating an input touch pattern, and <a2> is a diagram exemplarily illustrating a grouping process of sensing data related to the touch pattern.

Referring to FIG. 5, the server 100 may receive a touch pattern TP through the terminal screen SC of the user terminal 200. For example, the touch pattern TP may pass through a first point d1, a second point d2, a third point d3, and a fourth point d4. Accordingly, the touch pattern TP may pass through the first to fourth regions ar1 to ar4 respectively set for the first to fourth points d1 to d4. However, the present invention is not limited thereto, and it goes without saying that the touch pattern TP may have a shape that passes only an area set for a point without passing through a point.

Subsequently, the server 100 may receive sensing data for the touch pattern TP. The server 100 may receive sensing data for the first area ar1, the second area ar2, the third area ar3, and the fourth area ar4 in which the touch pattern TP is formed.

In this case, the sensing data may be sensing data generated in a preset time unit. For convenience of explanation, hereinafter, it is assumed that the preset time unit is 1 second.

The server 100 may receive the first sensing data T1 and the second sensing data T2 for the touch pattern TP passing through the first point d1 and the first area ar1 for 2 seconds. . In this case, the first sensing data T1 and the second sensing data T2 may be sensing data generated for 1 second, respectively. When the touch pattern TP is formed in an area outside the preset area, the server 100 may receive the third sensing data T3 and the fourth sensing data T4 for the area outside the area.

Subsequently, when the touch pattern TP is formed on the second point d2 and the second area ar2, the server 100 receives the fifth sensing data for the second point d2 and the second area ar2. T5) and sixth sensing data T6 may be received. Subsequently, the server 100 transmits the seventh sensing data to the sixteenth to the touch pattern TP passing through the third point d3, the third area ar3, the fourth point d4, and the fourth area ar4. Sensing data T7 to T16 may be received.

Subsequently, the server 100 may set a plurality of sensing data derived from a line connecting two adjacent points in the input touch pattern TP as one group.

In this case, the server 100 may set a start point and an end point for the received sensing data. The starting point may be temporally a first-order sensing data among sensing data received for two adjacent points, and the ending point may be a second-order sensing data. Subsequently, the server 100 may group the sensing data using the start point and the end point.

For example, the server 100 may set the first sensing data T1 initially received for the first area ar1 set around the first point d1 as a first starting point. Subsequently, the server 100 temporally receives the sixth sensing data for the second area ar2 set around the second point d2 closest to the first point d1 in the touch pattern TP. (T6) can be set as the first end point. The server 100 includes first sensing data T1 as a first starting point, sixth sensing data T6 as a first ending point, and first to sixth sensing data T1 received between the first starting point and the first ending point. ~T6) may be set as the first group G1.

In addition, the server 100 may set the fifth to eleventh sensing data T5 to T11 derived from a line connecting the second point d2 and the third point d3 as one group. That is, the server 100 sets the fifth sensing data T5 of the second point d2 and the second area ar2 as the second starting point, and the third point d3 and the third area ar3 The eleventh sensing data T11 may be set as the second end point. The server 100 may set the fifth to eleventh sensing data T5 to T11 as the second group G2.

In addition, the server 100 may set the ninth to sixteenth sensing data T9 to T16 derived from a line connecting the third point d3 and the fourth point d4 as the third group G3.

As a result, the number of groups grouped in the touch pattern TP may be smaller than the number of preset regions through which the touch pattern TP passes. In addition, sensing data overlapping with each other may exist in each adjacent group. For example, the fifth and sixth sensing data T5 and T6 may be commonly included in the first group G1 and the second group G2.

Subsequently, the server 100 may calculate a representative value for each parameter of the first group G1, the second group G2, and the third group G3. In this case, the parameter may be composed of at least two or more of touch time, touch position, touch pressure, touch size, data of a gyro sensor, or data of an acceleration sensor. In addition, the server 100 may calculate the average value, maximum value, minimum value, standard deviation, median value, RMS or RSS of each parameter to calculate a representative value.

Subsequently, the server 100 may generate a touch pattern profile for each of the first to third groups G1 to G3 using the representative value.

Hereinafter, a method of calculating the similarity using the generated touch pattern profile will be described.

6 is a diagram illustrating an example of the present invention for calculating a similarity in step S260 of FIG. 3. Here, <b1> of FIG. 6 represents sensing data of a previously registered customer registration pattern profile, and <b2> represents sensing data of a newly inputted touch pattern profile.

Referring to FIG. 6, the server 100 may associate a previously registered customer registration pattern profile with a touch pattern profile generated for a touch pattern input through the user terminal 200.

Subsequently, the server 100 may compare a representative value of sensing data for each group of a customer registration pattern profile registered in advance with a representative value of sensing data for each group of a touch pattern profile corresponding to each group.

Subsequently, the server 100 may calculate the similarity of the customer registration pattern profile and the touch pattern profile for each group. For example, the server 100 includes a representative value of the first registration group GR1 generated based _{on the sensing data X 11} to X _1n extracted between the first point and the second point of the customer registration pattern, The first similarity may be calculated using a representative value of the first group G1 generated based _{on the sensing data Y 11} to Y _1n extracted between the first point and the second point of the touch pattern.

In this case, the server 100 may calculate the first similarity by using a cosine similarity formula. For example, the server 100 may calculate the first degree of similarity using the following <Equation 1>.

[Equation 1]

Here, CS represents the cosine similarity, x represents the representative value of each group of the customer registration pattern profile, y represents the representative value of each group of the touch pattern profile, and n represents the parameter included in the sensing data of each group. Represents the number of

Subsequently, the server 100 calculates a second degree of similarity between the second registration group Gr2 and the second group G2 using the above-described method, and between the third registration group Gr3 and the third group G3. The third degree of similarity of can be calculated.

Subsequently, the server 100 may calculate an average similarity for the calculated plurality of similarities. That is, the server 100 may calculate an average degree of similarity for the first degree of similarity, the second degree of similarity, and the third degree of similarity. However, the embodiment using the above formula is only an example of the present invention, and it is obvious that various formulas capable of calculating the similarity can be applied.

FIG. 7 is a diagram illustrating another example of the present invention in which a similarity degree is calculated using a deep learning module in step S260 of FIG. 3. That is, the present invention may perform user authentication using the similarity calculated using the deep learning module, instead of the method of calculating the similarity using an equation.

Specifically, referring to FIG. 7, the deep learning module DM may receive a touch pattern profile and a customer registration pattern profile, and may provide a similarity between the touch pattern profile and the customer registration pattern profile as an output thereof.

At this time, the deep learning module (DM) receives a representative value for each group for the sensing data of the customer registration pattern profile created using a preset area and a representative value for each group for the sensing data of the currently input touch pattern profile. I can.

Subsequently, the deep learning module (DM) may derive a degree of similarity using an artificial neural network that has been pre-learned based on big data. The derived degree of similarity may be used to perform the above-described user identification or identification.

The deep learning module DM may perform machine learning on parameters input as learning factors. The deep learning module (DM) may train an artificial neural network using collected data (ie, positive data) of a corresponding customer input as a learning factor and data collected from other customers (ie, negative data). In this case, data and result data used for machine learning may be stored in the memory of the server 100.

In more detail, deep learning technology, a kind of machine learning, is to learn by going down to the deep level in multiple stages based on data.

Deep learning refers to a set of machine learning algorithms that extract core data from a plurality of data while increasing the level.

The deep learning module (DM) may use various known deep learning structures. For example, the deep learning module (DM) may use a structure such as a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN), a Deep Belief Network (DBN), and a Graph Neural Network (GNN).

Specifically, CNN (Convolutional Neural Network) is based on the assumption that a person recognizes an object based on the result of complex calculations after extracting the basic features of an object when a person recognizes an object. It is a replica model.

RNN (Recurrent Neural Network) is widely used for natural language processing, etc., and is an effective structure for processing time-series data that changes with the passage of time, and can construct an artificial neural network structure by stacking layers every moment.

DBN (Deep Belief Network) is a deep learning structure composed by stacking multiple layers of RBM (Restricted Boltzman Machine), a deep learning technique. When RBM (Restricted Boltzman Machine) learning is repeated to become a certain number of layers, a Deep Belief Network (DBN) having the corresponding number of layers may be formed.

GNN (Graphic Neural Network, hereinafter GNN) represents an artificial neural network structure implemented by deriving similarity and feature points between modeling data using modeling data modeled based on data mapped between specific parameters. .

Meanwhile, the artificial neural network learning of the deep learning module (DM) can be performed by adjusting the weight of the connection line between nodes (if necessary, adjusting the bias value) so that a desired output is produced for a given input. In addition, the artificial neural network may continuously update the weight value by learning. In addition, a method such as back propagation may be used for learning of the artificial neural network.

Meanwhile, an artificial neural network previously learned by machine learning may be mounted in the memory of the server 100.

The deep learning module DM may perform a machine learning-based improvement process recommendation operation using modeling data for the derived parameters as input data. In this case, both semi-supervised learning and supervised learning may be used as a machine learning method of the artificial neural network. In addition, the deep learning module DM may be controlled to automatically update an artificial neural network structure for outputting a touch pattern similarity after learning according to a setting.

Additionally, although not clearly shown in the drawings, in another embodiment of the present invention, the operation of the deep learning module DM may be performed in the server 100 or a separate cloud server (not shown).

In summary, the user authentication method of the present invention groups sensing data of the touch pattern by deriving a line connecting two adjacent points from a touch pattern input by a user, and a touch pattern generated based on the grouped sensing data. By calculating and comparing the similarity between the profile and the pre-registered customer registration pattern profile, it is possible to enhance the security of authentication compared to the method of comparing only the matching of the existing touch pattern. At the same time, since the same input interface as the method using the existing touch pattern is used, the inconvenience that the user has to adapt to the new authentication environment can be eliminated.

8 is a flowchart illustrating a user authentication method according to another embodiment of the present invention. Hereinafter, contents overlapping with the contents of the above-described embodiment of the present invention will be omitted and described.

Referring to FIG. 8, the server 100 extracts an initial touch pattern initially input through the user terminal 200 (S310 ).

Specifically, when a user performs a specific task (eg, membership registration of an application) through the user terminal 200, the server 100 may extract the user's touch pattern while performing the corresponding task.

In this case, the server 100 may extract the user's touch for each predetermined window size (d). Here, the window may move to the right every preset time (t) in a sliding window method.

Subsequently, based on the extracted first touch pattern, the server 100 generates a customer registration pattern profile (S320). The server 100 may group by receiving sensing data for an initial touch pattern.

In this case, the sensing data may be sorted and received in the order of generation time, and the server 100 may generate a profile of a customer registration pattern by grouping some of the sensing data received through a sliding window method. For example, the server 100 may set a plurality of sensing data included in a preset reference time unit as one group. A range of a plurality of groups to be set may be set using a reference time unit based on different starting points. Details about this will be described with reference to FIG. 9.

As another example, although not clearly shown in the drawing, the server 100 may divide the received sensing data into events of a predetermined type, and create a group to include sensing data corresponding to the reference number of events. . In this case, each group to be set may have a range set using a reference number of events based on a different starting point. Contents for grouping using events will be described in detail in FIGS. 10 and 11.

Subsequently, the server 100 may calculate a representative value for each parameter of the generated sensing data for each group, and generate a customer registration pattern profile based on the calculated representative value.

Subsequently, the server 100 extracts a touch pattern input through the user terminal 200 during the user's app activity (S330). In this case, the server 100 may extract a touch pattern in real time for all touches that occur while the user uses the application. The server 100 may generate a touch pattern profile by grouping sensing data for the extracted touch patterns and calculating a representative value for each group based on the sensing data for each group.

Subsequently, the server 100 receives a financial transaction request from the application (S340).

Subsequently, the server 100 calculates a similarity between the generated touch pattern profile and the previously generated customer registration pattern profile (S350). The server 100 may calculate a similarity for each group of the touch pattern profile and the corresponding customer registration pattern profile, and calculate an average value of the calculated similarity for each group as a final similarity.

Subsequently, the server 100 determines whether the calculated similarity is greater than the reference value (S360). When the calculated similarity is greater than a preset reference value, the server 100 may perform the requested financial transaction through the user terminal 200 (S371).

On the other hand, when the similarity of the touch pattern is less than the preset reference value, the server 100 may request additional authentication using the user terminal 200 (S373).

Hereinafter, a method of grouping sensing data using a reference time unit or a predetermined type of event will be described in detail.

FIG. 9 is a diagram illustrating an example of the present invention in which sensing data is grouped using a reference time unit in step S330 of FIG. 8.

Referring to FIG. 9, the server 100 may receive a plurality of sensing data for a user's touch pattern. In this case, the sensing data may be received after being sorted in the order of generation time.

Subsequently, the server 100 may generate a group by applying a reference time unit to the plurality of received sensing data. Each group may be created based on a different starting point, and a range may be set using a reference time unit. In this case, the server 100 may set all the received sensing data as respective starting points. However, the present invention is not limited thereto.

Hereinafter, for convenience of explanation, the reference time unit is 5 seconds, and the sensing data is generated once every second.

In this case, the server 100 may receive the first to n+2 sensing data T1 to Tn+2, which is a plurality of sensing data for the touch pattern.

Subsequently, the server 100 may extract sensing data for all touch patterns occurring within 5 seconds, which is a reference time unit, from the received sensing data.

Subsequently, the server 100 may set the first sensing data T1 generated in a time-sequential priority among sensing data generated within 5 seconds as a first starting point. The server 100 uses the first to fifth sensing data T1 to T5 generated for all touch patterns generated within 5 seconds, which is a reference time unit, including the first sensing data T1 as a first window W1. Can be grouped.

Subsequently, the server 100 sets the second sensing data T2 as a second starting point and sets the second to sixth sensing data T2 to T6 generated within 5 seconds as one group, and the second window W2 ) Can be created. In this case, the second to fifth sensing data T2 to T5 may be simultaneously included in the first window W1 and the second window W2.

Subsequently, the server 100 provides a third window W3 for the third to seventh sensing data T3 to T7, and a fourth window W4 for the fourth to eighth sensing data T4 to T8. An n-2th window W _n-2 may be generated for n-2 to n+1th sensing data T _n-2 to T _n+2.

That is, through the above-described method, the server 100 may group sensing data received in a preset reference time unit using a sliding window method.

10 and 11 are diagrams for explaining another example of the present invention in which sensing data is grouped using a gesture type in step S330 of FIG. 8. Here, FIG. 10 is a diagram illustrating a type of a predetermined event, and FIG. 11 is a diagram illustrating a method of grouping sensing data for events classified into a predetermined type.

Referring to FIG. 10, the server 100 may collect sensing data from the moment when the first touch occurs on the screen of the user terminal 200, and may terminate sensing data collection when the touch is terminated. In this case, the server 100 may classify the event type using basic touch information about a touch generated on the screen of the user terminal 200 or a type of sensing data collected for the touch.

For example, the server 100 may classify sensing data into event types of a finger touch operation (onDown; GT0), throwing (onFling; GT1), scrolling (onScroll; GT2) and single tap (onSingleTap; GT3). However, this is only an example, and it goes without saying that an additional event type may be added.

Specifically, when a touch event occurs for the first time in the user terminal 200, the server 100 may set the initial touch operation as a finger touch operation GT0. The finger touch operation GT0 may include an x-coordinate where a touch occurs, a y-coordinate where a touch occurs, a touch pressure, a touch size, data of a gyro sensor or data of an acceleration sensor. The finger touch operation GT0 may precede all events such as throwing (GT1), scrolling (GT2), and single tapping (GT3).

The throw GT1 may represent an event of sliding the screen of the user terminal 200 after the finger touch operation GT0 occurs. In this case, the throwing (GT1) operation may include the x-coordinate and y-coordinate, the touch pressure, the touch size, the speed of the x-coordinate and y-coordinate, data of a gyro sensor, or data of an acceleration sensor.

The scroll GT2 may indicate an event of scrolling a screen after a finger touch operation GT0. The scroll GT2 operation may include an x-coordinate and y-coordinate, a touch pressure, a touch size, a changed distance of the x-coordinate and y-coordinate, data of a gyro sensor, or data of an acceleration sensor. In this case, the server 100 may calculate the changed distance by using the distances of the x-coordinate and y-coordinate received at the position where the scroll GT2 is terminated in the x-coordinate and y-coordinate where the touch operation GT0 has occurred.

Subsequently, the single tap GT3 may be an event that occurs when the user removes a finger from the screen of the user terminal 200 after the finger touch operation GT0 event. At this time, since the single tap GT3 is in a state in which the touch is terminated, any data may not be included.

Meanwhile, referring to FIG. 11, the server 100 may receive data classified by event type. In this case, the server 100 may receive the event type and sensing data for the event type in the order of occurrence time.

Subsequently, the server 100 may set a group based on a reference number of events set in advance for a plurality of set sensing data.

For example, when the reference number is three, the server 100 may create a group by using sensing data for the first received finger touch operation. The server 100 sets the first sensing data R1 as a starting point, and includes the second sensing data R2 and the third sensing data R3 received immediately after receiving the first sensing data R1. Sensing data may be set as the first group. In addition, the server 100 may set the second sensing data R2 as a starting point and may set the second to fourth sensing data R2 to R4 as a second group.

Subsequently, the server 100 may calculate a representative value for each parameter of the sensing data for each group, and extract a touch pattern profile based on the calculated representative value.

Subsequently, the server 100 may extract an event of the same type in each generated group, and calculate a representative value of each parameter for sensing data included in the extracted event of the same type. For example, the server 100 may extract representative values for throw (GT1), scroll (GT2), and single tap (GT3) for each group.

Subsequently, the server 100 may connect the representative values extracted for each event type from each group into one vector. For example, the server 100 may connect representative values for throwing (GT1), scrolling (GT2), and single tapping (GT3) events extracted from the first to nth groups into one vector. Subsequently, the generated single vector may be included in the touch pattern profile.

Additionally, data of the finger touch operation GT0 may be included in an event immediately following and reflected in a representative value of the event. In addition, data of the last finger touch operation GT0 in the group may be discarded without being reflected in calculating the representative value. However, this is only an example, and the present invention is not limited thereto.

That is, through the above-described method, the server 100 may group the received sensing data by classifying the sensing data into a preset event type and using the reference number of events.

FIG. 12 is a diagram illustrating a method of calculating a similarity by using a deep learning module in step S350 of FIG. 8. Hereinafter, contents overlapping with the above-described contents will be omitted and described.

Specifically, referring to FIG. 12, the deep learning module DM may receive a touch pattern profile and a customer registration pattern profile, and output the similarity of a touch pattern between the touch pattern profile and the customer registration pattern profile as an output thereof.

At this time, the deep learning module (DM) is a representative value for each group for the sensing data of the customer registration pattern profile generated using a preset reference time unit, and a representative value for each group for the sensing data of the touch pattern profile input from the user. Can be input.

Meanwhile, when the touch pattern profile is generated using a preset event type, the touch pattern profile may be composed of a single vector including representative values for throw (GT1), scroll (GT2), and single tap (GT3) events. have. In this case, the customer registration pattern profile may also be configured in the same single vector form as the touch pattern profile.

Subsequently, the deep learning module DM may derive the similarity of the touch pattern by using the artificial neural network learned in advance based on the big data. The deep learning module DM may perform machine learning on parameters input as learning factors. The deep learning module (DM) may train an artificial neural network using collected data (ie, positive data) of a corresponding customer input as a learning factor and data collected from other customers (ie, negative data). In this case, data and result data used for machine learning may be stored in the memory of the server 100.

The artificial neural network learning of the deep learning module (DM) can be accomplished by adjusting the weight of the connection line between nodes (and adjusting the bias value if necessary) so that a desired output is produced for a given input. In addition, the artificial neural network may continuously update the weight value by learning. In addition, a method such as back propagation may be used for learning of the artificial neural network.

Meanwhile, an artificial neural network previously learned by machine learning may be mounted in the memory of the server 100.

The deep learning module DM may perform a machine learning-based improvement process recommendation operation using modeling data for the derived parameters as input data. In this case, both semi-supervised learning and supervised learning may be used as a machine learning method of the artificial neural network. In addition, the deep learning module DM may be controlled to automatically update an artificial neural network structure for outputting a touch pattern similarity after learning according to a setting.

Additionally, although not clearly shown in the drawings, in another embodiment of the present invention, the operation of the deep learning module DM may be performed in the server 100 or a separate cloud server (not shown). Hereinafter, a configuration of the deep learning module (DM) according to an embodiment of the present invention described above will be described.

13 is a diagram showing the configuration of a deep learning module used in some embodiments of the present invention. Here, the configuration of the deep learning module shown in FIG. 13 may be an embodiment applicable to the deep learning module (DM) described above with reference to FIGS. 7 and 12.

Referring to FIG. 13, the deep learning module (DM) is an input layer using a touch pattern profile and a customer registration pattern profile as an input node, and an output layer using a similarity between the touch pattern profile and the customer registration pattern profile as an output node. It includes (Output) and M hidden layers disposed between the input layer and the output layer.

Here, a weight may be set on an edge connecting nodes of each layer. The presence or absence of these weights or edges can be added, removed, or updated during the learning process. Accordingly, through the learning process, weights of nodes and edges disposed between k input nodes and i output nodes may be updated.

Before the deep learning module (DM) performs learning, all nodes and edges may be set to initial values. However, when information is accumulated and inputted, the weights of nodes and edges are changed, and parameters input as learning factors in this process (i.e., representative values for each parameter of each group of the touch pattern profile and registration pattern profile or single Vector) and a value assigned to the output node (ie, similarity between the touch pattern profile and the customer registration pattern profile) may be matched.

Additionally, when using a cloud server (not shown), the deep learning module DM may receive and process a large number of parameters. Therefore, the deep learning module (DM) can perform learning based on vast amounts of data.

The weights of nodes and edges between the input node and the output node constituting the deep learning module DM may be updated by a learning process of the deep learning module DM. In addition, it goes without saying that the parameters output from the deep learning module (DM) can be further extended to various data in addition to the similarity between the touch pattern profile and the customer registration pattern profile.

Subsequently, the server 100 may determine whether the touch pattern similarity is greater than a preset reference value based on the calculated similarity, and if it is greater than the preset reference value, the server 100 may provide a financial transaction requested from the user terminal 200.

Accordingly, in the user authentication method of the present invention, the sensing data extracted from the touch pattern input by the user are grouped using a reference time unit or a reference number of preset events, and a touch pattern generated based on the grouped sensing data By calculating and comparing the degree of similarity between the profile and the previously registered customer registration pattern profile, it is possible to improve the security of authentication.

In addition, since a separate pattern input step for authentication can be omitted, user convenience can be improved, and accuracy of authentication can be improved by analyzing the similarity using a touch pattern that the user normally inputs.

In addition, the user authentication method of the present invention calculates the similarity between the touch pattern profile calculated by the above-described method and the customer registration pattern profile using a pre-learned deep learning module, and authenticates the user using the calculated similarity, The accuracy and security of user authentication can be improved.

14 is a diagram illustrating a hardware implementation of a system that performs a user authentication method according to some embodiments of the present invention.

Referring to FIG. 14, the server 100 performing the user authentication method according to some embodiments of the present invention may be implemented as an electronic device 1000. The electronic device 1000 may include a controller 1010, an input/output device 1220 (I/O), a memory device 1230, an interface 1040, and a bus 1250. The controller 1010, the input/output device 1020, the memory device 1030, and/or the interface 1040 may be coupled to each other through the bus 1050. The bus 1050 corresponds to a path through which data is moved.

Specifically, the controller 1010 is a CPU (Central Processing Unit), MPU (Micro Processor Unit), MCU (Micro Controller Unit), GPU (Graphic Processing Unit), microprocessor, digital signal processor, microcontroller, application processor (AP , application processor), and at least one of logic elements capable of performing functions similar to these.

The input/output device 1020 may include at least one of a keypad, a keyboard, a touch screen, and a display device. The memory device 1030 may store data and/or programs.

The interface 1040 may perform a function of transmitting data to a communication network or receiving data from a communication network. The interface 1040 may be wired or wireless. For example, the interface 1040 may include an antenna or a wired/wireless transceiver. Although not shown, the memory device 1030 is an operation memory for improving the operation of the controller 1010 and may further include a high-speed DRAM and/or SRAM. The memory device 1030 may store a program or application therein.

The user terminal 200 includes a personal digital assistant (PDA), a portable computer, a web tablet, a wireless phone, a mobile phone, and a digital music player. music player), memory card, or any electronic product capable of transmitting and/or receiving information in a wireless environment.

Alternatively, the server 100 and the user terminal 200 according to embodiments of the present invention may be a system formed by connecting a plurality of electronic devices 1000 to each other through a network, respectively. In this case, each module or combination of modules may be implemented as the electronic device 1000. However, this embodiment is not limited thereto.

Additionally, the server 100 includes a workstation, a data center, an internet data center (IDC), a direct attached storage (DAS) system, a storage area network (SAN) system, and a network attached storage (NAS) system. It may be implemented as at least one of a system and a redundant array of inexpensive disks, or redundant array of independent disks (RAID) system, but the present embodiment is not limited thereto.

In addition, the server 100 may transmit data through a network using the user terminal 200. The network may include a network based on wired Internet technology, wireless Internet technology, and short-range communication technology. Wired Internet technology may include, for example, at least one of a local area network (LAN) and a wide area network (WAN).

Wireless Internet technologies include, for example, Wireless LAN (WLAN), Digital Living Network Alliance (DMNA), Wireless Broadband: Wibro, World Interoperability for Microwave Access (Wimax), and High Speed Downlink Packet (HSDPA). Access), High Speed Uplink Packet Access (HSUPA), IEEE 802.16, Long Term Evolution (LTE), Long Term Evolution-Advanced (LTE-A), Wireless Mobile Broadband Service (WMBS) And at least one of 5G New Radio (NR) technology. However, this embodiment is not limited thereto.

Short-range communication technologies include, for example, Bluetooth, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra-Wideband (UWB), ZigBee, and Near Field Communication: NFC), Ultra Sound Communication (USC), Visible Light Communication (VLC), Wi-Fi, Wi-Fi Direct, at least one of 5G NR (New Radio) It may include. However, this embodiment is not limited thereto.

The server 100 communicating through a network may comply with a technical standard and a standard communication method for mobile communication. For example, standard communication methods are GSM (Global System for Mobile communication), CDMA (Code Division Multi Access), CDMA2000 (Code Division Multi Access 2000), EV-DO (Enhanced Voice-Data Optimized or Enhanced Voice-Data Only). , WCDMA (Wideband CDMA), HSDPA (High Speed Downlink Packet Access), HSUPA (High Speed Uplink Packet Access), LTE (Long Term Evolution), LTEA (Long Term Evolution-Advanced) and at least one of 5G NR (New Radio) It may include. However, this embodiment is not limited thereto.

The above description is merely illustrative of the technical idea of the present embodiment, and those of ordinary skill in the technical field to which the present embodiment pertains will be able to make various modifications and variations without departing from the essential characteristics of the present embodiment. Accordingly, the present embodiments are not intended to limit the technical idea of the present embodiment, but to explain the technical idea, and the scope of the technical idea of the present embodiment is not limited by these embodiments. The scope of protection of this embodiment should be interpreted by the claims below, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present embodiment.

Claims (14)

In the user authentication method performed in an authentication server linked to a user terminal,
Receiving sensing data for a touch pattern input on the screen of the user terminal;
Grouping the plurality of received sensing data;
Calculating a representative value for each parameter of the sensing data for a plurality of sensing data included in each grouped group;
Generating a touch pattern profile for each group using the calculated representative values for each parameter;
Calculating a similarity between the generated touch pattern profile and a customer registration pattern profile previously stored and previously stored from the user terminal; And
Including the step of determining whether to authenticate the service requested by the user terminal based on the calculated similarity,
In each of the grouped groups, the first group and the second group adjacent to each other include sensing data generated in common point regions
User authentication method.
The method of claim 1,
The touch pattern is formed to pass through a plurality of point areas respectively set in a peripheral area of the corresponding point, centering on a plurality of preset points,
The grouping includes setting a plurality of sensing data derived from a line connecting two adjacent dot regions within the touch pattern as one group.
User authentication method.
The method of claim 2,
The touch pattern is formed to pass through first to third regions respectively set around first to third points set in advance,
The grouping step,
First sensing data for the first area set around the first point as a first starting point, and second sensing data for the second area set around the second point as a first end point, Set sensing data between the first start point and the first end point as the first group,
Sensing data between the second starting point and the second ending point by using the second sensing data as a second starting point and third sensing data for the third area set around the third point as a second ending point Comprising setting to the second group
User authentication method.
The method of claim 3,
The sensing data is generated in a preset time unit,
The first group and the second group commonly include the second sensing data for the second area set around the second point, and the first and second groups receive sensing data overlapping each other. Inclusive
User authentication method.
The method of claim 2,
The number of groups grouped in the touch pattern is smaller than the number of regions through which the touch pattern passes.
User authentication method.
The method of claim 2,
The step of determining whether to authenticate,
If the similarity is higher than a preset reference value and the touch pattern matches a preset registration pattern, approving the authentication of the service.
User authentication method.
In the user authentication method performed in an authentication server linked to a user terminal,
Receiving sensing data for a touch pattern input on the screen of the user terminal;
Grouping the plurality of received sensing data;
Calculating a representative value for each parameter of the sensing data for a plurality of sensing data included in each grouped group;
Generating a touch pattern profile for each group using the calculated representative values for each parameter;
Calculating a similarity between the generated touch pattern profile and a customer registration pattern profile previously stored and previously stored from the user terminal; And
Including the step of determining whether to authenticate the service requested by the user terminal based on the calculated similarity,
The touch pattern includes a plurality of sensing data arranged in the order of generation time,
The grouping step,
A group is created using the sensing data included in the preset reference time unit,
Each of the groups is set with a range based on a different starting point using the reference time unit,
Among the generated plurality of groups, a plurality of groups adjacent to each other in chronological order include the same sensing data in common.
User authentication method.
In the user authentication method performed in an authentication server linked to a user terminal,
Receiving sensing data for a touch pattern input on the screen of the user terminal;
Grouping the plurality of received sensing data;
Calculating a representative value for each parameter of the sensing data for a plurality of sensing data included in each grouped group;
Generating a touch pattern profile for each group using the calculated representative values for each parameter;
Calculating a similarity between the generated touch pattern profile and a customer registration pattern profile previously stored and previously stored from the user terminal; And
Including the step of determining whether to authenticate the service requested by the user terminal based on the calculated similarity,
The touch pattern includes a plurality of sensing data arranged in the order of generation time,
Each of the sensing data is divided into a predetermined type of event,
The grouping step,
Create a group to include sensing data corresponding to the predetermined reference number of events,
Each of the groups is created by setting a range based on a different starting point using the reference number of events,
Among the generated plurality of groups, a plurality of groups adjacent to each other in chronological order include the same sensing data in common.
User authentication method.
The method of claim 8,
The event is classified into any one type of throwing (onfling), scrolling (onscroll) and single tapping (onsingleTap),
Each of the above events is classified by a finger touch operation (onDown).
User authentication method.
The method of claim 8,
The step of calculating a representative value for each parameter,
Extracting the event of the same type within the same group,
For the sensing data included in the extracted event of the same type, including calculating a representative value for each parameter,
In the step of generating a touch pattern profile for each group,
The touch pattern profile includes one vector generated by connecting each event type composed of the representative value.
User authentication method.
The method of claim 1, 7 or 8,
The parameters of the sensing data are composed of at least two or more of touch time, touch position, touch pressure, touch size, data of a gyro sensor, or data of an acceleration sensor,
The step of calculating the representative value includes calculating an average value, a maximum value, a minimum value, a standard deviation, a median value, a root mean square (RMS), or a root sum square (RSS) of each parameter.
User authentication method.
The method of claim 1, 7 or 8,
The step of calculating the similarity,
Comprising each group-specific similarity to the group-specific touch pattern profile derived from the touch pattern and a pre-stored customer registration pattern profile corresponding thereto, and calculating an average of the calculated plurality of similarities,
The step of determining whether to authenticate,
If the calculated average is higher than a preset reference value, it is determined that authentication is successful,
If the calculated average is less than a preset reference value, including requesting additional authentication
User authentication method.
The method of claim 12,
The step of calculating the similarity includes calculating the similarity for each group using a deep learning module that has been learned in advance,
The deep learning module,
An input layer using the touch pattern profile and the customer registration pattern profile as input nodes,
An output layer having the similarity as an output node,
Including one or more hidden layers disposed between the input layer and the output layer,
The weights of nodes and edges between the input node and the output node are updated by the learning process of the deep learning module.
User authentication method.
In the user authentication method performed in an authentication server linked to a user terminal,
Receiving sensing data for a touch pattern input on the screen of the user terminal;
Grouping the plurality of received sensing data;
Generating a touch pattern profile for each group based on a plurality of sensing data included in each grouped group;
Calculating a similarity between the generated touch pattern profile and a customer registration pattern profile previously stored and previously stored from the user terminal; And
Including the step of determining whether to authenticate the service requested by the user terminal based on the calculated similarity,
The grouping step,
A plurality of sensing data derived from a line connecting two adjacent points in the touch pattern is set as one group, or
Including setting the sensing data corresponding to the preset reference time unit or the preset reference number of events as one group,
Among the generated plurality of groups, a plurality of groups adjacent to each other in chronological order include the same sensing data in common.
User authentication method.

Images