TWI783689B - Method for authenticating user identity based on touch operation - Google Patents

Abstract

A method for authenticating user identity based on touch operation includes a training stage and an authentication stage. The training stage includes: generating, by the a touch interface, a plurality of training touch parameters; and generating, by a processor ,a training heat map according to the plurality of training touch parameters. The authentication stage includes: generating, by the touch interface, a plurality of testing touch parameters; generating, by the processor, a test heat map according to the plurality of testing touch parameters; comparing, by the processor, the testing heat map with the training heat map to generate an error map; and generateing, by the processor, one of an authentication pass signal and an authentication fail signal according to the error map.

Description

Method for authenticating user identity based on touch operation

The present invention relates to user identity authentication, in particular to a method for authenticating user identity based on touch operation.

User authentication allows a computing device (such as a laptop or smartphone) to authenticate the identity of the person attempting to use the computing device, with a binary "pass" or "fail" result. Passing the authentication means that the user is indeed who he/she claims to be, while an authentication failure means that the user is not who he/she claims to be.

Passwords are a common mechanism for user authentication. However, if the password is stolen, the user who obtained the password will gain control of the account. Even if the password has not been compromised, when an authenticated user temporarily leaves the computing device he/she has logged into with the password, other unauthorized users can gain access to the computing device. This situation is called insider attack ( insider attack).

When passwords are at risk of being stolen, biometric authentication can be used to solve this problem. The biometric authentication method compares the user's biometrics with the records in the database. Typical biometrics technologies include fingerprints, faces, retinas, iris, and voice. Since biometric authentication uses unique biometrics for authentication, it is difficult to copy or steal.

However, biometrics are highly sensitive and private. A potential problem with biometric authentication is that biometric authentication systems may be used to perform functions beyond their original intent, a condition known as function creep. For example, the biometric authentication system installed in the workplace was originally used to prevent ordinary employees from entering confidential places. However, system administrators can also use this system to track all the places that individual employees have visited without prior notification to employees. This in turn violates the privacy rights of employees.

In addition, user identity authentication by password or biometric feature is usually one-time, and even if repeated authentication is required, there will be a certain period of time interval, otherwise it will bring additional troubles to the user. In other words, the existing user identity authentication methods are not continuous and non-intrusive.

In view of this, the present invention proposes a method for authenticating a user's identity based on a touch operation. This method achieves continuous and non-intrusive authentication of the user by observing the user's touch operation mode. If the touch pattern of the new user is too different from that of the old user, the new user will be judged as an impostor, and its access will be prohibited. The method proposed by the present invention can be called "TouchPrint", because the use dynamics of the touch operation can authenticate the user's identity like a fingerprint.

A method for authenticating a user's identity based on a touch operation according to an embodiment of the present invention includes: a training phase, including: a touch interface generating a plurality of training touch parameters; and a processor according to the training touch parameters generating a training heat map; and a certification stage, including: the touch interface generates a plurality of test touch parameters; the processor generates a test heat map according to the test touch parameters; the processor compares the test heat map and the training heat map to generate an error map; and the processor generates one of an authentication pass signal and an authentication failure signal according to the error map.

A method for authenticating a user's identity based on a touch operation according to an embodiment of the present invention includes: a training phase, including: a touch interface generating a plurality of training touch parameters; and a processor according to the training touch parameters. parameter training a neural network model; and a certification stage, including: the touch interface generates a plurality of test touch parameters; the processor inputs the test touch parameters to the neural network model to generate a prediction value; and The processor calculates an error value between the predicted value and an actual value to generate one of an authentication pass signal and an authentication failure signal.

A method for authenticating a user's identity based on a touch operation according to an embodiment of the present invention includes: a training phase, including: a touch interface generating a plurality of training touch parameters; and a processor according to the training touch parameters generating a training heat map and training a neural network model; and a certification stage, including: the touch interface generates a plurality of test touch parameters; the processor generates a test heat map according to the test touch parameters; the processing The device compares the test heat map and the training heat map to generate an error map; the processor calculates a first error value according to the error map; the processor inputs the test touch parameters to the neural network model to generate a predicted value; and the processor generates one of an authentication pass signal and an authentication failure signal based on the first error value and a second error value, wherein the second error value is associated with the predicted value and an actual value .

In summary, the method for authenticating users based on touch operations proposed by the present invention has the following effects:

1. Continuously and non-intrusively monitor and authenticate current users;

2. It is data-driven and adaptable, and can adapt to users' ever-changing modes of touch operation; and

3. Increase the frequency of user authentication, at times when existing user authentication mechanisms (such as passwords or biometrics) do not.

It is worth noting that the method proposed by the present invention does not mean to replace the existing user identity authentication mechanism, but to supplement and enhance the security of the existing mechanism to avoid internal attacks. In other words, an impostor can be detected even when the computing device is unlocked.

The above description of the disclosure and the following description of the implementation are used to demonstrate and explain the spirit and principle of the present invention, and to provide further explanation of the patent application scope of the present invention.

The detailed features and characteristics of the present invention are described in detail below in the implementation mode, and its content is enough to enable any person familiar with the relevant art to understand the technical content of the present invention and implement it accordingly, and according to the content disclosed in this specification, the scope of the patent application and the drawings , anyone who is familiar with the related art can easily understand the ideas and features related to the present invention. The following examples are to further describe the concept of the present invention in detail, but not to limit the scope of the present invention in any way.

The method for authenticating users based on touch operations proposed by the present invention includes nine embodiments, of which the first embodiment, the second embodiment, and the third embodiment use the heat map created by touch operations as the authentication basis Compared with the first embodiment, the second and third embodiments add an authentication basis update mechanism, but the execution sequence of the update mechanism in the second and third embodiments is different. The fourth embodiment, the fifth embodiment, and the sixth embodiment use the neural network model established by the touch operation as the authentication basis. Compared with the fourth embodiment, the fifth and sixth embodiments add an update mechanism for the authentication basis, but the fifth embodiment 6. The execution sequence of the update mechanism in the sixth embodiment is different. The seventh embodiment, the eighth embodiment, and the ninth embodiment use heat maps and neural networks as the authentication basis. Compared with the seventh embodiment, the eighth and ninth embodiments add an update mechanism for the authentication basis, but the The execution sequence of the update mechanism in the eighth and ninth embodiments is different.

FIG. 1 is a flow chart of a method for authenticating a user based on a touch operation according to the first embodiment of the present invention, wherein: a touch operation is defined as any operation performed by the user in a specific area, and the type of touch operation may include movement , pressing (and releasing later), and scrolling, but the present invention is not limited thereto. As shown in FIG. 1 , the method for authenticating a user based on a touch operation includes a training phase S1 and an authentication phase S2 , and is applicable to a computing device with a touch interface and a processor. Computing devices such as notebook computers or smart phones, and touch interfaces such as physical touch panels or touch screens. In one embodiment, the touch interface can also be composed of a physical device and a display interface, such as a mouse and a screen displaying the mouse cursor, a motion sensor and a projector for projecting the sensed motion onto a surface, a camera device and an application program that converts captured actions into movement tracks, but the present invention is not limited to the above-mentioned examples, any software or hardware device that allows the user to control the movement of a marker (such as a cursor) within a specified range, are applicable to the present invention.

Fig. 2 is a detailed flowchart of the steps in Fig. 1. As shown in Fig. 2, the training phase S1 includes steps S11 and S13, and the authentication phase S2 includes steps S21, S23, S25, and S27. The training stage S1 is used to collect training data of multi-touch operations generated by a specific user through the touch interface, and summarize the behavior pattern of the specific user. The authentication stage S2 is used to collect the test data of the multi-touch operation generated by the current user through the touch interface, and determine whether the behavior pattern corresponding to these data is similar to the behavior pattern generated in the training stage S1.

In the training stage S1, step S11 is "the touch interface generates multiple training touch parameters", step S13 is "the processor generates a training heat map according to these training touch parameters", and step S21 is "the touch interface generates multiple test Touch parameters", step S23 is "the processor generates a test heat map according to these test touch parameters", step S25 is "the processor compares the test heat map and the training heat map to generate an error map", step S27 is "the processor one of an authentication pass signal and an authentication failure signal is generated according to the error map".

In steps S11 and S21 , the content of each of the training touch parameters and the test touch parameters includes: touch time point, touch position and operation type. Through the driver program of the touch interface or the software development kit (Software Development Kit, SDK) of the operating system, the processor can access the log file (Log) of the touch operation as shown in Table 1 below.

Table 1, log file of touch operation touch point operation type touch position 2020-02-18 10:13:29.735389 move to (1284, 568) 2020-02-18 10:13:29.735389 move to (1284, 567) 2020-02-18 10:13:29.766641 press to (1284, 567) 2020-02-18 10:13:29.844739 released on (1284, 567) 2020-02-18 10:13:29.922846 move to (1283, 567) … … … … 2020-02-18 10:13:59.256981 scroll to (721, 454)

In steps S13 and S23, each of the training heatmap (heatmap) and the test heatmap includes at least one of a position heatmap and a speed heatmap, and the position heatmap is used to reflect the cumulative number of touch positions, and processing The detector generates multiple location heat maps corresponding to different operation types, as shown in Fig. 3, Fig. 4 and Fig. 5. FIG. 3 is a position heat map of two different users based on moving operations, FIG. 4 is a position heat map of these two users based on pressing operations, and FIG. 5 is a position heat map of these two users based on scrolling operations.

FIG. 6 is a position heat map and a speed heat map based on a mobile operation of the same user, wherein the speed heat map is used to reflect the direction and distance of the mobile operation. In one embodiment, the processor first calculates a plurality of velocity vectors and then generates a velocity heat map, wherein each velocity vector is composed of two consecutive moving operations. In the speed heat map on the right side of Figure 6, the horizontal axis is the direction of the movement operation, and the unit is radian; the vertical axis is the distance of the movement operation, and the unit is pixel. For example: the movement operation takes 3 milliseconds to move from coordinates (0, 0) to coordinates (6, 12), then the speed of this movement operation is 2√5 pixels per millisecond; and the component of this movement operation on the X axis is 2 Pixels, the component on the Y axis is 4 pixels, the angle of this movement operation is about 1.10714872 degrees (rad).

Step S25 is mainly to apply template matching (template matching) technology. Specifically, the processor compares the training heatmap and the test heatmap to generate an error map, wherein the types of the training heatmap and the test heatmap used for comparison must be the same, for example, both are position heatmaps or both are is the velocity heatmap. In a first example of step S25, the accuracy of the comparison is, for example, at least one of a pixel scale and a block scale. The pixel-scale comparison is for example: calculating the difference between the grayscale values of the pixels at the same position in the training heatmap and the test heatmap; and the block-scale comparison is for example: calculating the difference between the training heatmap and the test heatmap Structural similarity index measure (SSIM). In the second example of step S25, the processor first cuts the training heatmap and the test heatmap into multiple feature spaces (eigenspaces), performs a rotation operation for each feature space, and then adopts the method of the first example, A feature space in the training heatmap is randomly selected for comparison with a feature space in the test heatmap. In the second example, the same touch pattern of the same user at different touch positions can be found.

In an extended embodiment based on FIG. 2, the processor further collects the operating program (process) currently executed by the computing device in the training phase S1 and the authentication phase S2, such as the application program corresponding to the foreground window in the operating system (such as Microsoft Word , Google Chrome, etc.), and generate corresponding training heatmaps and test heatmaps according to different operating procedures. Therefore, when the processor compares the training heatmap and the test heatmap, the types of the two must be the same, and the corresponding operating procedures must also be the same.

It is worth noting that, according to various touch types or various operating procedures, the processor may generate multiple training heat maps in step S13; in step 23, the processor generates multiple test heat maps corresponding to the multiple training heat maps ; Therefore, in step S25, the processor generates a plurality of error maps, and the present invention does not limit the upper limit of the number of error maps.

In step S27, the processor calculates one or more error values according to an error map, and then compares each error value with a corresponding threshold value. If the error value exceeding the specified number is greater than the threshold value, the processor generates an authentication failure signal; otherwise, if the error value exceeding the specified number is not greater than the threshold value, the processor generates an authentication passing signal.

Fig. 7 is a flowchart of a method for authenticating a user based on a touch operation according to the second embodiment of the present invention. Compared with the first embodiment, the second embodiment mainly adds an update stage S3'. Fig. 8 is a detailed flowchart of the steps in Fig. 7, as can be seen from Fig. 8, the training stage S1' of the second embodiment is more step S12' added than the training stage S1 of the first embodiment, and the authentication stage S2' of the second embodiment A step S22' is added to the authentication stage S2 of the first embodiment. In the following, only the newly added parts of the second embodiment will be described, and the same steps as those of the first embodiment will not be repeated.

Step S12' is "the processor judges that the collection amount of these training touch parameters is greater than the first threshold", and step S22' is "the processor judges that the collection amount of these test touch parameters is greater than the test threshold". The collection amount may be at least one of a time interval (for example, touch parameters collected within 72 hours) and a number of parameters (for example, one hundred thousand touch parameters). After the touch parameters are collected in step S11' and step S21', the judging mechanism of step S12' and step S22' needs to be met before proceeding to step S13' and step S23' respectively.

Please refer to FIG. 7 and FIG. 8, in the second embodiment, the updating phase S3' is located between the training phase S1' and the authentication phase S2'. Step S31' is "the touch interface generates a plurality of new touch parameters", step S32' is "the processor judges that the collection amount of these new touch parameters is greater than the second threshold", and step S33' is "the processor Training the touch parameters to generate a new training heat map", and step S34' is "the processor updates the training heat map according to the new training heat map". Basically, steps S31'~S33' in the update stage S3' are the same as steps S11'~S13' in the training stage S1'. In addition, in the second embodiment, the values of the first threshold in step S12', the second threshold in step S32', and the test threshold in step S22' are not particularly limited.

In step S34', the processor calculates the difference between the new training heat map and the training heat map. The training parameters generate a new training heatmap (different from the training heatmap of step S1'). When the difference is not less than the update threshold, the processor replaces the original training heatmap generated in step S13' with the new training heatmap generated in step S33'.

Overall, since the user's touch pattern may change with time or the current operating procedures, therefore, the second embodiment first collects a set of touch parameters to initialize a reference training heat map, and then collects Another set of touch parameters to update the original training heatmap.

9 is a flow chart of a method for authenticating a user based on a touch operation according to a third embodiment of the present invention. The training phase S1' and the authentication phase S2' of the third embodiment are basically the same as those of the second embodiment. In other words, the third embodiment is equivalent to adding steps S12' and S22' in the first embodiment, and adding an update stage S3' after the authentication stage S2'. In other words, the third embodiment is based on the second embodiment, and moves the execution order of the update phase S3' to after the authentication phase S2'.

The application method of the third embodiment is as follows: a user generates its exclusive training heat map in the training stage S1', and the same user can continue to verify the accuracy of the training heat map judgment through the test heat map in the authentication stage S2', and then Feedback the accuracy rate to the update stage S3', thereby modifying the second threshold value in step S32'. For example, if in the authentication phase S2', the correct rate in step S27' is less than a certain threshold, the processor will increase the second threshold in step S32' to collect more new touch parameters, so as to improve the next time Correct rate when performing authentication phase S2'.

In the second embodiment and the third embodiment, the update stage S3' can also update the touch parameters in a regular manner to reflect the change of the user's touch mode over time, and the update method is as follows: Method 1:

為更新後的觸控參數，

為步驟S31’所述的新觸控參數，

為步驟S11’所述的觸控參數，

及

是權重。 in

is the updated touch parameter,

is the new touch parameter described in step S31',

is the touch parameter described in step S11',

and

is the weight.

FIG. 10 is a flow chart of a method for authenticating users based on touch operations according to the fourth embodiment of the present invention. FIG. 11 is a detailed flow chart of the steps in FIG. 10, wherein step T11 is "generation of multiple training touch parameters on the touch interface" , step T13 is "the processor trains the neural network model according to these training touch parameters", step T21 is "the touch interface generates a plurality of test touch parameters", and step T23 is "the processor inputs these test touch parameters to the neural network Network model to generate predicted value", step T25 is "the processor calculates the error value between the predicted value and the actual value to generate one of an authentication pass signal and an authentication failure signal".

The main difference between the fourth embodiment and the first embodiment is the step T13 in the training phase T1 and the steps T23, T25 in the authentication phase T2, so the details of these different steps T13, T23, T25 are only mentioned below, as for the fourth The same steps as those in the first embodiment will not be repeated.

In step T13, the processor converts multiple training touch parameters into time series (time series), and then uses the time series as the input layer of the neural network model for training, and obtains a prediction model for predicting subsequent time series . The neural network model is, for example, a Long Short-Term Memory (LSTM) model. In addition, multiple predictive models can be trained according to touch parameters of different touch types.

The generation methods of time series may include but not limited to the three methods listed below:

1. The time series consists of the time and location of each mobile operation.

2. The time series consists of a start time (or end time) and a velocity vector, where the velocity vector is calculated by the processor taking out two consecutive multiple moving operations at fixed time intervals. Because the velocity vector is a two-dimensional vector, this time series is a multivariate time series.

3. The time series consists of one or more centroids of each training heatmap and the time points corresponding to each training heatmap. For example, multiple training touch parameters are divided into multiple groups in chronological order, and a training heat map is generated for each group of training touch parameters, and then the centroid of each training heat map is calculated, such as by using K-mean Algorithms (K-means).

In step T23, the processor may input a set of test touch signals to the predictive model generated in step T13 to obtain a set of predicted values, and the set of predicted values may include one or more test touch signals.

The first example of step T25 adopts "H Nguyen, Kim Phuc Tran, S Thomassey, M Hamad. Forecasting and Anomaly Detection approaches using LSTM and LSTM Autoencoder techniques with the applications in Supply Chain Management. International Journal of Information Management, Elsevier, 2020 The Anomaly detection mechanism of the time series introduced by .

The description of the anomaly detection mechanism is as follows: The next touch parameter, such as the position, can be predicted through the LSTM model. If the actual location is too far from the predicted location, this actual location is considered an anomaly. By using an auto-encoder (auto-encoder), such as an LSTM encoder or an image auto-encoder, the template (norm) as an authentication standard can be learned. The auto-encoder is equivalent to a feature extractor, regardless of the auto-encoder. Inputs are general images, heatmaps, time series or time series/heatmaps of multivariate statistics. Taking the example of a location heatmap, the template represents the commonly touched areas of the touchpad and is represented as a low-dimensional or embedded space. In the certification stage T2, the test heatmap is fed into the autoencoder and reconstructed. Once the difference between the test heatmap and the reconstructed test heatmap is too large, this test heatmap is considered an anomaly.

The second example of step T25 is as follows: the processor obtains another set of test touch signals whose time sequence is after the set of test touch signals described in step T23, sets it as a set of actual values, and then calculates the set of actual values. The error value between each of the values and the corresponding predicted value. The processor compares each error value with its corresponding threshold value. If the error value exceeding the specified number is greater than the threshold value, the processor generates an authentication failure signal; otherwise, if the error value exceeding the specified number is not greater than the threshold value, the processor generates an authentication pass signal.

Fig. 12 is a flowchart of a method for authenticating a user based on a touch operation according to a fifth embodiment of the present invention. Compared with the fourth embodiment, the fifth embodiment mainly adds an update stage T3'. FIG. 13 is a detailed flowchart of the steps in FIG. 12 . It can be seen from Fig. 13 that the training stage T1' of the fifth embodiment has more steps T12' than the training stage T1 of the fourth embodiment, and the authentication stage T2' of the fifth embodiment has more steps than the authentication stage T2 of the fourth embodiment T22', the same steps in the fifth embodiment and the fourth embodiment will not be repeated below. In addition, the update stage T3' of the fifth embodiment has a similar structure to the update stage S3' of the second embodiment, the difference between them is: step T33' is "the processor updates the neural network according to the new training touch parameters road model", on the other hand step S33' is "the processor updates the training heat map according to the new training touch parameters". To sum up, these two embodiments update different authentication reference standards according to the new training touch parameters.

14 is a flow chart of a method for authenticating a user based on a touch operation according to a sixth embodiment of the present invention. The training stage T1' and the authentication stage T2' of the sixth embodiment are basically the same as those of the fifth embodiment. In other words, the sixth embodiment is based on the fifth embodiment, and moves the execution sequence of the update phase T3' to after the authentication phase T2'.

FIG. 15 is a flowchart of a method for authenticating a user based on a touch operation according to the seventh embodiment of the present invention, including a training phase U1 and an authentication phase U2. FIG. 16 is a detailed flowchart of the steps in FIG. 15 . The training phase U1 includes steps U11, U13, and the authentication phase U2 includes steps U21, U23, U24, U25, U26, U27.

Step U11 is "the touch interface generates multiple training touch parameters", step U13 is "the processor generates a training heat map and training neural network model according to these training touch parameters", and step U21 is "the touch interface generates multiple Test touch parameters", step U23 is "processor generates test heat map according to these test touch parameters", step U24 is "processor compares test heat map and training heat map to generate error map", step U25 is "processing The device calculates the first error value according to the error map", step U26 is "the processor inputs these test touch parameters to the neural network model to generate a predicted value", step U27 is "the processor calculates the first error value and the second error value according to the Generate one of an authentication pass signal and an authentication failure signal". It can be known from the above that the seventh embodiment integrates the first embodiment and the fourth embodiment, and at the same time adopts the training heat map and the neural network model for user authentication.

Fig. 17 is a flowchart of a method for authenticating a user based on a touch operation according to the eighth embodiment of the present invention. Compared with the seventh embodiment, the eighth embodiment mainly adds an update stage U3'. Fig. 18A and Fig. 18B are detailed flow charts of the steps in Fig. 17. It can be seen from Fig. 18A and Fig. 18B that the training phase U1' of the eighth embodiment is more step U12' than the training phase U1 of the seventh embodiment, and the eighth embodiment In the authentication phase U2' of the example, a step U22' is added to the authentication phase U2 of the seventh embodiment. The same steps in the eighth embodiment and the seventh embodiment will not be repeated below. In addition, the eighth embodiment is equivalent to the integration of the second embodiment and the fifth embodiment, in other words, steps U23', U24' are equivalent to steps S23', S25', and step U26' is equivalent to T26'. Based on integration requirements, the eighth embodiment adds steps U25' and U27'. In step U25', the processor calculates a first error value according to the error map. In step U27', the processor generates one of an authentication pass signal and an authentication failure signal according to the first error value and the second error value, wherein the second error value is associated with the predicted value obtained in step S26' and obtained from step U23 'Retrieve the actual value. For details about the predicted value, the actual value and the second error value, please refer to the part about the calculation of the error value in step T27. For example, the processor calculates a weighted average of the first error value and the second error value, and judges whether the weighted average is greater than a threshold value, so as to decide to output an authentication pass signal or an authentication failure signal.

In addition, the update stage U3' of the eighth embodiment is equivalent to integrating S3' of the second embodiment and T3' of the fifth embodiment. In other words, in step U33', the processor not only generates a new training heat map according to the new training touch parameters, but also updates the neural network model according to the new training touch parameters. And in step U34', the processor updates the training heatmap according to the new training heatmap. To sum up, processors use two different certification reference standards at the same time.

Fig. 19 is a flowchart of a method for authenticating a user based on a touch operation according to the ninth embodiment of the present invention. The training phase U1' and the authentication phase U2' of the ninth embodiment are basically the same as those of the eighth embodiment. In other words, the ninth embodiment is based on the eighth embodiment, and the execution sequence of the update phase U3' is moved after the authentication phase U2'.

In the seventh embodiment, the eighth embodiment and the ninth embodiment, the present invention not only uses the template matching technology to compare the training heat map and the test heat map, but also adds artificial intelligence (AI) The technology provides additional comparison criteria for the user's unique touch recognition.

In summary, the method for authenticating users based on touch operations proposed by the present invention has the following effects:

1. Continuously and non-intrusively monitor and authenticate current users;

2. It is data-driven and adaptable, and can adapt to users' ever-changing modes of touch operation; and

3. Increase the frequency of user authentication, at times when existing user authentication mechanisms (such as passwords or biometrics) do not.

It is worth noting that the method proposed by the present invention does not mean to replace the existing user identity authentication mechanism, but to supplement and enhance the security of the existing mechanism to avoid internal attacks. In other words, an impostor can be detected even when the computing device is unlocked.

Although the present invention is disclosed by the aforementioned embodiments, they are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, all changes and modifications are within the scope of patent protection of the present invention. For the scope of protection defined by the present invention, please refer to the appended scope of patent application.

S1, S1’, T1, T1’, U1, U1’: training phase S2, S2’, T2, T2’, U2, U2’: authentication phase S3, S3’, T3, T3’, U3, U3’: update phase S11, S13, S21, S23, S25, S27: steps S11’, S12’, S13’, S21’, S22’, S23’, S25’, S27’, S31, S32’S33’, S34: steps T11, T13, T21, T23, T25: steps T11’, T12’, T13’, T21’, T22’, T23’, T25’, T31, T32’T33’: steps U11, U13, U21, U23, U24, U25, U26, U27: Steps U11’, U12’, U13’, U21’, U22’, U23’, U24, U25’, U26, U27’: steps U31, U32'U33', U34: steps

FIG. 1 is a flowchart of a method for authenticating a user based on a touch operation according to a first embodiment of the present invention; Fig. 2 is the detailed flowchart of step among Fig. 1; Figure 3 is a location heat map of two different users based on the mobile operation location; Fig. 4 is a position heat map based on pressing operations of two different users; Fig. 5 is a location heat map of two different users based on the scrolling operation; Fig. 6 is the position heat map and speed heat map based on the mobile operation of the same user; 7 is a flow chart of a method for authenticating a user based on a touch operation according to a second embodiment of the present invention; Fig. 8 is a detailed flowchart of steps among Fig. 7; 9 is a flow chart of a method for authenticating a user based on a touch operation according to a third embodiment of the present invention; 10 is a flow chart of a method for authenticating a user based on a touch operation according to a fourth embodiment of the present invention; Figure 11 is a detailed flowchart of the steps in Figure 10; FIG. 12 is a flowchart of a method for authenticating a user based on a touch operation according to a fifth embodiment of the present invention; Figure 13 is a detailed flowchart of the steps in Figure 12; 14 is a flowchart of a method for authenticating a user based on a touch operation according to the sixth embodiment of the present invention; 15 is a flow chart of a method for authenticating a user based on a touch operation according to the seventh embodiment of the present invention; Figure 16 is a detailed flowchart of the steps in Figure 15; 17 is a flow chart of a method for authenticating a user based on a touch operation according to the eighth embodiment of the present invention; 18A and 18B are detailed flowcharts of the steps in FIG. 17; and FIG. 19 is a flowchart of a method for authenticating a user based on a touch operation according to the ninth embodiment of the present invention.

S1: training stage

S2: Authentication stage

S11~S13, S21~S27: steps

Claims (7)

A method for authenticating a user's identity based on a touch operation, comprising: a training phase, including: a touch interface generating a plurality of training touch parameters; and a processor generating a training heat map according to the training touch parameters; and A certification stage, including: the touch interface generates a plurality of test touch parameters; the processor generates a test heat map according to the test touch parameters; the processor compares the test heat map and the training heat map to generate an error graph; and the processor generates one of an authentication pass signal and an authentication failure signal according to the error graph, wherein the contents of each of the training touch parameters and the test touch parameters include: touch Control time point, touch position and operation type, the operation type includes: at least one of moving, pressing and scrolling. The method for authenticating user identity based on touch operation as described in claim 1, wherein each of the training heat map and the test heat map includes at least one of a position heat map and a speed heat map, and the position heat map is used for Reflecting the accumulated times of touch positions, the speed heat map is used to reflect the direction and distance of a moving operation. The method for authenticating user identity based on touch operation as described in claim 1, wherein: the training phase further includes: before the processor generates the training heat map, the processor determines that the collection amount of the training touch parameters is greater than the first threshold; The authentication stage further includes: before the processor generates the test heat map, the processor determines that the collection amount of the test touch parameters is greater than a test threshold; and the method further includes an update stage, the update stage includes: The touch interface generates a plurality of new touch parameters; the processor determines that the collection amount of the new touch parameters is greater than a second threshold; the processor generates a new training heat map according to the new training touch parameters; and The processor updates the training heatmap according to the new training heatmap. A method for authenticating user identity based on touch operation, comprising: a training phase, comprising: a touch interface generates a plurality of training touch parameters; and a processor trains a neural network model according to the training touch parameters ; and a certification stage, comprising: the touch interface generates a plurality of test touch parameters; the processor inputs the test touch parameters to the neural network model to generate a predicted value; and the processor calculates the predicted value and an error value of an actual value to generate one of an authentication pass signal and an authentication failure signal, wherein the contents of each of the training touch parameters and the test touch parameters include: touch time point . Touch position and operation type, the operation type includes: at least one of moving, pressing and scrolling. The method for authenticating user identity based on touch operation as described in claim 4, wherein: The training phase further includes: before the processor generates the neural network model, the processor judges that the collection amount of the training touch parameters is greater than a first threshold; the authentication phase further includes: inputting the Before testing the touch parameters to the neural network model, the processor determines that the collection amount of the test touch parameters is greater than the test threshold; and the method further includes an update stage, the update stage includes: the touch interface generates A plurality of new touch parameters; the processor determines that the collection amount of the new touch parameters is greater than a second threshold; and the processor updates the neural network model according to the new training touch parameters. A method for authenticating a user's identity based on a touch operation, comprising: a training phase, including: a touch interface generating a plurality of training touch parameters; and a processor generating a training heat map and training based on the training touch parameters A neural network model; and a certification stage, including: the touch interface generates a plurality of test touch parameters; the processor generates a test heat map according to the test touch parameters; the processor compares the test heat map and the training heat map to generate an error map; the processor calculates a first error value according to the error map; the processor inputs the test touch parameters to the neural network model to generate a predicted value; and the processing The device generates one of an authentication pass signal and an authentication failure signal according to the first error value and a second error value, wherein the second error value is associated with the predicted value and an actual value, Each of the training heat map and the test heat map includes at least one of a position heat map and a speed heat map, the position heat map is used to reflect the cumulative number of touch positions, and the speed heat map is used to reflect a movement Direction and distance of operation. The method for authenticating user identity based on touch operation as described in claim 6, wherein: the training stage further includes: before the processor generates the training heat map and trains the neural network model, the processor determines the training The collection amount of touch parameters is greater than the first threshold; the certification stage further includes: before the processor generates the test heat map and trains the neural network model, the processor determines that the collection amount of these test touch parameters is greater than A threshold value is tested; and the method further includes an update stage, the update stage includes: the touch interface generates a plurality of new touch parameters; the processor determines that the collection amount of these new touch parameters is greater than a second threshold value; The processor generates a new training heat map and updates the neural network model according to the new training touch parameters; and the processor updates the training heat map according to the new training heat map.

Images