CN108804902A - A method of the electronics Freehandhand-drawing safety verification based on deep learning model - Google Patents
A method of the electronics Freehandhand-drawing safety verification based on deep learning model Download PDFInfo
- Publication number
- CN108804902A CN108804902A CN201810569539.7A CN201810569539A CN108804902A CN 108804902 A CN108804902 A CN 108804902A CN 201810569539 A CN201810569539 A CN 201810569539A CN 108804902 A CN108804902 A CN 108804902A
- Authority
- CN
- China
- Prior art keywords
- user
- information
- model
- freehandhand
- present
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2221/00—Indexing scheme relating to security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F2221/21—Indexing scheme relating to G06F21/00 and subgroups addressing additional information or applications relating to security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F2221/2133—Verifying human interaction, e.g., Captcha
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- User Interface Of Digital Computer (AREA)
- Collating Specific Patterns (AREA)
Abstract
The method of the invention discloses a kind of electronics Freehandhand-drawing safety verification based on deep learning model, the present invention by machine learning are that differentiating for future is extracted and the Freehandhand-drawing mode and Freehandhand-drawing feature of storage user in verification process in user's registration.During differentiating extraction, the present invention tracks plotting speed and track quickly carries out subscriber authentication by the hand drawing pattern and gesture of analysis user.The present invention identifies that finger is touched by high-precision and carries out the authentication of user in the drawing track of control panel and Touch Screen to guarantee network security.The present invention by machine learning in user's registration be future differentiate verification process extraction and storage user Freehandhand-drawing mode and feature.The present invention to ensure network security, can improve the safety of network information, and then improve the intelligent of system by the hand drawing pattern and gesture path for verifying user to avoid the leakage of userspersonal information.
Description
Technical field
The invention belongs to the technical fields of multi-media information security verification, and in particular to a kind of based on deep learning model
The method of electronics Freehandhand-drawing safety verification.
Background technology
With the fast development of internet, internetwork machine people detection is widely used in various websites, in order to
Prevent someone from using rogue program frequent visit website, website generally can judge that operator is people using the mode of identifying code
Class or program.Such technical solution is relatively more at present, common are identifying code, click verification picture etc..But with shifting
Dynamic internet is popularized, and handset size is small, inputs troublesome characteristic and so that current scheme is not very applicable.With computer
Technology constantly exploitation and perfect, user can use the terminal devices such as computer, mobile phone play, do shopping, work on the net,
User need to only input registered account, account number cipher and identifying code and delivery operation can be completed in an application terminal,
Since the payment information inputted is too simple, being easy to be identified by other people during input leads to information leakage, reduces
The safety of network information thereby reduces the intelligent of system.
Invention content
The method of the purpose of the present invention is to provide a kind of electronics Freehandhand-drawing safety verification based on deep learning model, this hair
The bright Freehandhand-drawing mode and hand for differentiating extraction and storage user in verification process by machine learning in user's registration for future
Paint feature;The present invention is identified to screen asking for robot manipulation by the information of hand drawing pattern and gesture path to user
Topic, at the same the present invention by verify user hand drawing pattern and gesture path can to avoid the leakage of userspersonal information, to
It has ensured network security, has improved the safety of network information, and then improved the intelligent of system.
The present invention is achieved through the following technical solutions:A kind of electronics Freehandhand-drawing safety verification based on deep learning model
Method, mainly include the following steps that:
Step S101:User registers in systems, and personal information is inputted on touch screen or control panel, and system extracts user
Hand drawing pattern and gesture path information and be sent in model;The model verifies the identity of user, if verification
Successful then model sends the information to succeed in registration by the hand drawing pattern of user and the data input database of gesture path;
Step S102:When user logs on system, then user needs to input personal information on the touchscreen, and system extraction is used
The hand drawing pattern at family and the information of gesture path are simultaneously sent in model;The model first verifies the identity of user,
If being proved to be successful, model carries out the information registered in the information and date library of the hand drawing pattern of user's checking and gesture path
It compares, if information comparison is consistent, sends the information being proved to be successful.
In order to which the present invention is better achieved, further, if model inspection is to the user's registered in the step S101
Identity is robot, then the information of system output registration failure;If model inspection to the user registered identity as true man, be
The information that system output is succeeded in registration.
In order to which the present invention is better achieved, further, if the identity of model inspection to user are in the step S102
When robot, then the information of system output authentication failed;If the identity of model inspection to the user of verification are true man, model
Further the information registered in the information and date library of the hand drawing pattern of user's checking and gesture path is compared.
The present invention extracts and stores user's in user's registration by machine learning for differentiating for future in verification process
Freehandhand-drawing mode and Freehandhand-drawing feature.During differentiating extraction, the present invention passes through the hand drawing pattern and gesture for analyzing user, tracking
Plotting speed and track quickly carry out subscriber authentication.The present invention identifies that finger is touched in control panel and touched by high-precision
The drawing track for controlling screen carries out the authentication of user to guarantee network security.The present invention is noted by machine learning in user
It is the Freehandhand-drawing mode and feature for differentiating verification process extraction and storage user in future when volume.
In user's registration, typing information in systems is needed, system extracts the hand drawing pattern and gesture path of user
The information of the essential information of user, hand drawing pattern and gesture path is simultaneously sent to model by information, at this time body of the model to user
Part is verified, and detection user is machine or true man;Go out drawing and gesture if user is true man by user if model inspection
Feature vector record in the database, then export the result of detection and terminate to verify, export the information to succeed in registration;If model
It detects that user is machine, then export the result of detection and terminates to verify, export the information of registration failure.
When user reuses, user's validation information, the hand drawing pattern and gesture rail of system records user are needed
Mark is simultaneously sent to model, and model verifies user identity at this time, and detection user is machine or true man;If model inspection arrives
When user is robot, then exports the result of detection and terminate to verify, export the information of authentication failed;If model inspection is to user
When being true man, then model is further by the hand drawing pattern received and gesture path data and the hand drawing pattern and hand in database
Verification is compared in gesture track data, if verify data information is consistent, exports the result of detection and terminates to verify, output verification
Otherwise successful information exports the information of authentication failed.
VelocityTracker be one tracking touch event sliding speed help class, for realizing flinging with
And other similar gestures.Its principle is that touch event MotionEvent objects are passed to VelocityTracker
Then addMovement (MotionEvent) method is analyzed the displacement that MotionEvent objects occur in unit interval class and is come
Calculating speed.Android.view.VelocityTracker is mainly with tracking touchscreen events(Flinging events and other
Gestures gesture events)Rate.Motion event are added to addMovement (MotionEvent) functions
In VelocityTracker class examples.It is laterally and perpendicular that you can use getXVelocity () or getXVelocity () to obtain
To rate to rate when, but please first call computeCurrentVelocity (int) to initialize using before them
The unit of rate.
When you need to track the speed of touchscreen events, obtained using obtain () method
One instance objects of VelocityTracker classes.In onTouchEvent call back functions, addMovement is used
(MotionEvent) current moving event is passed to VelocityTracker objects by function.It uses
ComputeCurrentVelocity (int units) functions calculate current speed, using getXVelocity (),
GetYVelocity () function obtains current speed.
Code is very simple, we can find out the pseudo- instantaneous velocity during move, in this way when doing many controls
Can all use, for example, system Launcher paging, ScrollView sliding etc., can be calculated according to speed at this time
Retarded motion etc. after ACTION_UP.
Android sdk are we provided GestureDetector(Gesture:Gesture Detector:Identification)Class,
By this class, we can identify many gestures, mainly be completed by his onTouchEvent (event) method
The identification of different gestures.Although he can identify gesture, it is different how gesture will be handled, it should be available to programmer's reality
Existing.
This class of GestureDetector externally provides two interfaces and an outer category:Interface:
OnGestureListener, OnDoubleTapListener;Inner classes:SimpleOnGestureListener.
This outer category is the integrated of all functions in two interfaces in fact, it contain it is all in the two interfaces must
It the function that need realize and has all rewritten, but all method bodies are all empty;Difference is:Such is static
Class, programmer can inherit this class in outside, rewrite the gesture processing method of the inside.
Using touch screen is touched, the touch screen can record the hand drawing pattern and gesture path of user.The model is tested
The method of user identity is demonstrate,proved by number of patent application is 2016103538829, the applying date is 2016.05.25 Chinese invention
Patent disclosure, and be not the improvement of the present invention, so it will not be repeated.
This patent of invention is different from other electronic handwritten signature systems, is identifying the same of common electronic handwritten signature
When, it can identify all hand drawing patterns and gesture, it is more safer than common password authentification and screen gesture unlock verification.
Moreover, quickly identifying subscriber authentication by the plotting speed and trace for tracking user(It is the verification of people or computer).
This is other verification modes(Password is arranged, identifying code extraction and electronic handwritten signature etc.)It is not easily achieved.
Beneficial effects of the present invention:
The present invention by machine learning in user's registration be future differentiate verification process in extract and storage user Freehandhand-drawing
Mode and Freehandhand-drawing feature.During differentiating extraction, the present invention passes through the hand drawing pattern and gesture for analyzing user, and tracking is drawn
Speed and track quickly carry out subscriber authentication.The present invention identifies that finger is touched in control panel and touch screen by high-precision
The drawing track of curtain carries out the authentication of user to guarantee network security.The present invention is by machine learning in user's registration
For the Freehandhand-drawing mode and feature for differentiating verification process extraction and storage user in future.
The present invention by machine learning in user's registration be future differentiate verification process extraction and storage user hand
Paint mode and feature;The present invention is identified to screen robot behaviour by the information of hand drawing pattern and gesture path to user
The problem of making, while the present invention can letting out to avoid userspersonal information by the hand drawing pattern and gesture path for verifying user
Dew, to ensure network security, improves the safety of network information, and then improve the intelligent of system.
Description of the drawings
Fig. 1 is the flow diagram of user's registration;
Fig. 2 is the flow diagram of user's checking.
Specific implementation mode
Embodiment 1:
A method of the electronics Freehandhand-drawing safety verification based on deep learning model mainly includes the following steps that:
Step S101:User registers in systems, and personal information is inputted on touch screen or control panel, and system extracts user
Hand drawing pattern and gesture path information and be sent in model;The model verifies the identity of user, if verification
Successful then model sends the information to succeed in registration by the hand drawing pattern of user and the data input database of gesture path;
Step S102:When user logs on system, then user needs to input personal information on the touchscreen, and system extraction is used
The hand drawing pattern at family and the information of gesture path are simultaneously sent in model;The model first verifies the identity of user,
If being proved to be successful, model carries out the information registered in the information and date library of the hand drawing pattern of user's checking and gesture path
It compares, if information comparison is consistent, sends the information being proved to be successful.
In use, model is screened to the identity information of user first to the present invention, judges to register user's
Operation is true man or robot manipulation, if model inspection is to being that true man operate, system can establish data to the information of user
Library, the information of hand drawing pattern and gesture path to user's typing;When user logs on system, then model is first to user
Identity screened, judgement be true man operation or machine operation, if model inspection to be true man operate when, system can be right
Information in the hand drawing pattern of user's checking typing and the information and date library of gesture path is matched, if information matches one
It causes, then the information reminding that system output is proved to be successful, otherwise exports the information reminding of authentication failed.
The present invention extracts and stores user's in user's registration by machine learning for differentiating for future in verification process
Freehandhand-drawing mode and Freehandhand-drawing feature.During differentiating extraction, the present invention passes through the hand drawing pattern and gesture for analyzing user, tracking
Plotting speed and track quickly carry out subscriber authentication.The present invention identifies that finger is touched in control panel and touched by high-precision
The drawing track for controlling screen carries out the authentication of user to guarantee network security.The present invention is noted by machine learning in user
It is the Freehandhand-drawing mode and feature for differentiating verification process extraction and storage user in future when volume.
The present invention by machine learning in user's registration be future differentiate verification process extraction and storage user hand
Paint mode and feature;The present invention is identified to screen robot behaviour by the information of hand drawing pattern and gesture path to user
The problem of making, while the present invention can letting out to avoid userspersonal information by the hand drawing pattern and gesture path for verifying user
Dew, to ensure network security, improves the safety of network information, and then improve the intelligent of system.
Embodiment 2:
The present embodiment is advanced optimized on the basis of embodiment 1, as shown in Figure 1 and Figure 2, if model in the step S101
The identity of the user registered is detected as robot, then the information of system output registration failure;If model inspection to registration use
The identity at family is true man, the then information that system output succeeds in registration;If identity of the model inspection to user in the step S102
For robot when, then the information of system output authentication failed;If the identity of model inspection to the user of verification are true man, mould
Further the information registered in the information and date library of the hand drawing pattern of user's checking and gesture path is compared for type.
In user's registration, typing information in systems is needed, system extracts the hand drawing pattern and gesture path of user
The information of the essential information of user, hand drawing pattern and gesture path is simultaneously sent to model by information, at this time body of the model to user
Part is verified, and detection user is machine or true man;Go out drawing and gesture if user is true man by user if model inspection
Feature vector record in the database, then export the result of detection and terminate to verify, export the information to succeed in registration;If model
It detects that user is machine, then export the result of detection and terminates to verify, export the information of registration failure.
When user reuses, user's validation information, the hand drawing pattern and gesture rail of system records user are needed
Mark is simultaneously sent to model, and model verifies user identity at this time, and detection user is machine or true man;If model inspection arrives
When user is robot, then exports the result of detection and terminate to verify, export the information of authentication failed;If model inspection is to user
When being true man, then model is further by the hand drawing pattern received and gesture path data and the hand drawing pattern and hand in database
Verification is compared in gesture track data, if verify data information is consistent, exports the result of detection and terminates to verify, output verification
Otherwise successful information exports the information of authentication failed.
The present invention by machine learning in user's registration be future differentiate verification process extraction and storage user hand
Paint mode and feature;The present invention is identified to screen robot behaviour by the information of hand drawing pattern and gesture path to user
The problem of making, while the present invention can letting out to avoid userspersonal information by the hand drawing pattern and gesture path for verifying user
Dew, to ensure network security, improves the safety of network information, and then improve the intelligent of system.
The other parts of the present embodiment are same as Example 1, and so it will not be repeated.
The above is only presently preferred embodiments of the present invention, not does limitation in any form to the present invention, it is every according to
According to the technical spirit of the present invention to any simple modification, equivalent variations made by above example, the protection of the present invention is each fallen within
Within the scope of.
Claims (3)
1. a kind of method of the electronics Freehandhand-drawing safety verification based on deep learning model, which is characterized in that include mainly following step
Suddenly:
Step S101:User registers in systems, and personal information is inputted on touch screen or control panel, and system extracts user
Hand drawing pattern and gesture path information and be sent in model;The model verifies the identity of user, if verification
Successful then model sends the information to succeed in registration by the hand drawing pattern of user and the data input database of gesture path;
Step S102:When user logs on system, then user needs to input personal information on the touchscreen, and system extraction is used
The hand drawing pattern at family and the information of gesture path are simultaneously sent in model;The model first verifies the identity of user,
If being proved to be successful, model carries out the information registered in the information and date library of the hand drawing pattern of user's checking and gesture path
It compares, if information comparison is consistent, sends the information being proved to be successful.
2. a kind of method of electronics Freehandhand-drawing safety verification based on deep learning model according to claim 1, feature
Be, if in the step S101 model inspection to the identity of the user registered as robot, system output registration failure
Information;If model inspection to the user registered identity as true man, information that system output succeeds in registration.
3. a kind of method of electronics Freehandhand-drawing safety verification based on deep learning model according to claim 2, feature
It is, if the identity of model inspection to user are robot in the step S102, the information of system output authentication failed;
If the identity of model inspection to the user of verification are true man, model is further by the hand drawing pattern of user's checking and gesture
The information registered in the information and date library of track is compared.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810569539.7A CN108804902A (en) | 2018-06-05 | 2018-06-05 | A method of the electronics Freehandhand-drawing safety verification based on deep learning model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810569539.7A CN108804902A (en) | 2018-06-05 | 2018-06-05 | A method of the electronics Freehandhand-drawing safety verification based on deep learning model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108804902A true CN108804902A (en) | 2018-11-13 |
Family
ID=64088575
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810569539.7A Pending CN108804902A (en) | 2018-06-05 | 2018-06-05 | A method of the electronics Freehandhand-drawing safety verification based on deep learning model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108804902A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109918883A (en) * | 2019-02-22 | 2019-06-21 | 袁精侠 | A kind of auth method of the biocompatibility characteristics based on Freehandhand-drawing track |
CN110349461A (en) * | 2019-06-11 | 2019-10-18 | 北京光年无限科技有限公司 | Education and entertainment combination method and system based on children special-purpose smart machine |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104618100A (en) * | 2013-12-23 | 2015-05-13 | 腾讯科技(深圳)有限公司 | Identity authentication method, method for paying based on terminal, terminal and server |
CN104992085A (en) * | 2015-06-13 | 2015-10-21 | 东莞市微模式软件有限公司 | Method and device for human body in-vivo detection based on touch trace tracking |
CN106066959A (en) * | 2016-05-25 | 2016-11-02 | 北京比邻弘科科技有限公司 | A kind of method and device of bot access detection |
CN108011868A (en) * | 2017-11-09 | 2018-05-08 | 武汉极意网络科技有限公司 | One kind slides verification method and mobile terminal |
-
2018
- 2018-06-05 CN CN201810569539.7A patent/CN108804902A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104618100A (en) * | 2013-12-23 | 2015-05-13 | 腾讯科技(深圳)有限公司 | Identity authentication method, method for paying based on terminal, terminal and server |
CN104992085A (en) * | 2015-06-13 | 2015-10-21 | 东莞市微模式软件有限公司 | Method and device for human body in-vivo detection based on touch trace tracking |
CN106066959A (en) * | 2016-05-25 | 2016-11-02 | 北京比邻弘科科技有限公司 | A kind of method and device of bot access detection |
CN108011868A (en) * | 2017-11-09 | 2018-05-08 | 武汉极意网络科技有限公司 | One kind slides verification method and mobile terminal |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109918883A (en) * | 2019-02-22 | 2019-06-21 | 袁精侠 | A kind of auth method of the biocompatibility characteristics based on Freehandhand-drawing track |
CN109918883B (en) * | 2019-02-22 | 2022-06-10 | 袁精侠 | Identity verification method of biological characteristics based on hand-drawn track |
CN110349461A (en) * | 2019-06-11 | 2019-10-18 | 北京光年无限科技有限公司 | Education and entertainment combination method and system based on children special-purpose smart machine |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11736478B2 (en) | Device, system, and method of user authentication based on user-specific characteristics of task performance | |
CN105740678B (en) | Electronic apparatus and its operating method | |
US10476873B2 (en) | Device, system, and method of password-less user authentication and password-less detection of user identity | |
US20160226865A1 (en) | Motion based authentication systems and methods | |
CN102216935B (en) | Apparatus and method for inputting password using game | |
US20150154392A1 (en) | Secure face authentication with liveness detection for mobile | |
CN105335641A (en) | Identity authentication method and device based on fingerprint identification | |
US9202035B1 (en) | User authentication based on biometric handwriting aspects of a handwritten code | |
Koong et al. | A user authentication scheme using physiological and behavioral biometrics for multitouch devices | |
CN103714457B (en) | Method for verifying transaction | |
US20210287221A1 (en) | Systems and methods for active signature detection | |
CN104809402A (en) | Fuzzy display method for information and terminal | |
CN105929974A (en) | Password input management method and mobile terminal | |
CN108804902A (en) | A method of the electronics Freehandhand-drawing safety verification based on deep learning model | |
CN107707738A (en) | A kind of face identification method and mobile terminal | |
CN107295016A (en) | Auth method and device | |
CN105207783B (en) | A kind of User reliability authentication method and device | |
Ali et al. | User behaviour-based mobile authentication system | |
Feng et al. | Security after login: Identity change detection on smartphones using sensor fusion | |
WO2020190165A1 (en) | Method and system for identifying a user from cursor movement trajectory | |
CN105653993A (en) | Password inputting method, apparatus and electronic device | |
Shuwandy et al. | BAWS3TS: Browsing authentication web-based smartphone using 3D touchscreen sensor | |
US11860998B2 (en) | Emulator detection through user interactions | |
CN110766837A (en) | Control method and device for passing equipment, machine readable medium and equipment | |
CN108460260A (en) | The method of detecting system, fingerprint sensor and related finger touching certification |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181113 |