WO2019153587A1 - 用户身份验证方法、装置、计算机设备和存储介质 - Google Patents

用户身份验证方法、装置、计算机设备和存储介质 Download PDF

Info

Publication number
WO2019153587A1
WO2019153587A1 PCT/CN2018/089058 CN2018089058W WO2019153587A1 WO 2019153587 A1 WO2019153587 A1 WO 2019153587A1 CN 2018089058 W CN2018089058 W CN 2018089058W WO 2019153587 A1 WO2019153587 A1 WO 2019153587A1
Authority
WO
WIPO (PCT)
Prior art keywords
typing
behavior model
model
input device
text information
Prior art date
Application number
PCT/CN2018/089058
Other languages
English (en)
French (fr)
Inventor
黄创茗
Original Assignee
平安科技(深圳)有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Publication of WO2019153587A1 publication Critical patent/WO2019153587A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00Network architectures or network communication protocols for network security
    • H04L63/08Network architectures or network communication protocols for network security for authentication of entities
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31User authentication
    • G06F21/316User authentication by observing the pattern of computer usage, e.g. typical user behaviour

Definitions

  • the application relates to a user identity verification method, device, computer device and storage medium.
  • a user identity verification method, apparatus, computer device, and storage medium capable of improving verification efficiency are provided.
  • a user authentication method includes:
  • a user identity verification device includes:
  • the information collecting module is configured to collect text information entered by the terminal and an entry interval of characters in the text information
  • a model generating module configured to generate a first typing behavior model according to the text information and the entry interval time
  • a template reading module for reading a typing habit feature template
  • the model generation module is further configured to convert the text information into a typing habit model by using the typing habit feature template;
  • the model comparison module is configured to compare the typing habit model with the first typing behavior model to obtain a user identity verification result.
  • a computer device comprising a memory and one or more processors having stored therein computer readable instructions, the computer readable instructions being executable by the processor to cause the one or more processors to execute The following steps:
  • One or more non-volatile storage media storing computer readable instructions, when executed by one or more processors, cause one or more processors to perform the following steps:
  • FIG. 1 is an application scenario diagram of a user identity verification method according to one or more embodiments
  • FIG. 2 is a schematic flow chart of a user identity verification method according to one or more embodiments
  • FIG. 3 is a schematic flow chart of a user identity verification method in another embodiment
  • FIG. 4 is a block diagram of a user identity verification device in accordance with one or more embodiments.
  • Figure 5 is a block diagram of a user identity verification apparatus in another embodiment
  • FIG. 6 is a block diagram of a user identity verification device in accordance with one or more embodiments.
  • FIG. 7 is a block diagram of a user identity verification apparatus in another embodiment
  • FIG. 8 is a block diagram of a user identity verification device in accordance with one or more embodiments.
  • Figure 9 is a block diagram of a user identity verification apparatus in another embodiment.
  • Figure 10 is a block diagram of a computer device in accordance with one or more embodiments.
  • Terminal 110 communicates with server 120 over a network over a network.
  • the server 120 collects the text input information entered by the terminal 110 and the entry interval time of the characters in the text information, and generates a first typing behavior model according to the text information and the entry interval time.
  • the server 120 also converts the text information into a typing habit model by using a typing habit feature template, and compares the typing habit model with the first typing behavior model to obtain a user identity verification result.
  • the terminal 110 can be, but is not limited to, various personal computers, notebook computers, smart phones, tablets, and portable wearable devices, and the server 120 can be implemented by a stand-alone server or a server cluster composed of a plurality of servers.
  • a user identity verification method is provided.
  • the method is applied to the server in FIG. 1 as an example, and includes the following steps:
  • S202 Collect text information entered by the terminal and an entry interval of characters in the text information.
  • the terminal can be the terminal in FIG.
  • the text information is a text file that records text information.
  • the text information may specifically be content that is expressed in a different arrangement by different characters.
  • the terminal enters the text information, and specifically, the text information is input through the input device on the text input box on the page provided by the server, and the text information is saved in the text information.
  • a character is the basic information of a unit of glyph, a glyph unit, or a symbol. Characters can be letters, numbers, words, symbols, and so on.
  • the entry interval is the time period between when the two characters are entered.
  • the server monitors the text box on the web page via a JavaScript script sent to the web page of the terminal.
  • the server detects that the terminal enters a character in the text box, it collects the characters entered by the terminal and the time point at which the terminal inputs the character. After the terminal enters the character, all the characters are saved in the text information, and the entry interval time of the adjacent two characters in the text information and the entry time of the text information are calculated according to the time point of the terminal input character.
  • a user identity verification method is also applied to the terminal. After triggering the instruction to input the text information, the terminal monitors the input device connected to the terminal in real time, and collects characters input by the input device and input characters input by the device. At the time point, the characters input by the input device are converted into characters recorded in the text information according to the input method stored on the terminal. After the character entry in the text information is completed, the entry interval of the characters in the text information is generated according to the character input by the input device and the time point at which the input device inputs the character.
  • the typing behavior model may specifically be a tool for explaining the corresponding typing behavior of the text information.
  • the typing behavior model can be either a mathematical formula or a chart.
  • Typing is the act of entering textual information through an input device.
  • the typing behavior may be a behavior of tapping a key on a physical keyboard, such as a user keystroke, or by touching a touch panel or operating a mouse to trigger the behavior of a virtual key on the virtual keyboard.
  • the server sequentially records the intervals according to the order of the characters in the text information, and generates an entry interval time sequence according to the sequentially arranged entry intervals.
  • the server generates a first typing behavior model according to the recording interval time series.
  • a user identity verification method is further applied to the terminal, and the terminal generates an input interval time sequence according to the character input by the input device and the time point of the input device inputting the character, and according to the text information and the corresponding entry interval time. Generate a sequence of entry intervals. The terminal generates a first typing behavior model according to the input interval time series and the input interval time series.
  • the typing habit feature template is a template for recording the characteristics of typing habits.
  • Typing habits are a collection of features of typing behavior. Specifically, if there is a typing habit generated according to a typing behavior of a user, the text information entered according to the typing habit matches the typing behavior of the user entering the text information.
  • the server obtains the user identity from the terminal, and reads the corresponding typing habit feature template according to the user identity.
  • the server also detects the number of retraining times of the read typing habit feature template and the time of the last retraining. When detecting the number of retraining times and/or the time of the last retraining exceeds a preset threshold, the input text is sent to the terminal multiple times. Instruction for information.
  • the typing habit model is a typing behavior model that is simulated according to the typing habit feature template. Specifically, the text information is converted into a typing habit model by using a typing habit feature template, and the characters matching the order of the characters in the text information may be selected in the typing habit feature template, for example, according to the order of the characters in the text information. Intervals.
  • the server selects the corresponding interval time according to the order of the characters in the text information, and arranges the corresponding interval time according to the arrangement order of the characters, and generates the typing according to the arranged interval time.
  • Custom model
  • a typing behavior model can be generated according to the collected text information and the recording interval time, and the typing of different users is different due to different typing habits of each user.
  • the behavior will not be the same.
  • the typing habit feature template extracted from the database can reflect the user's typing habits, and the text information is converted into a typing habit model through the typing habit feature template. Therefore, the typing habit model can also reflect the user's typing habits. By comparing the typing habit model with the typing behavior model, it can be compared whether the corresponding typing behavior of the typing behavior model matches the typing habits corresponding to the typing habit model.
  • the security of verification can be ensured by comparing whether typing behavior and typing habits match.
  • the text information that needs to be input is easily recognized by the user, and thus it is not easy to repeatedly verify due to the recognition error, thereby greatly reducing the time taken for the user to verify the identity, thereby improving the efficiency of the authentication.
  • the method after collecting the text information entered by the terminal, the method further includes: identifying a type of the input device for inputting the text information to the terminal; and when identifying the type of the input device, transmitting the type to the terminal Monitoring the monitoring instruction of the input device of the type; receiving the input device triggering information collected and fed back by the terminal according to the monitoring instruction; after step S208, the method further comprises: generating a second typing behavior model according to the input device triggering information; The typing habit model, the first typing behavior model and the second typing behavior model obtain the user identity verification result.
  • An input device is a device that inputs data and information to a computer device.
  • the input device may specifically be at least one of a keyboard, a mouse, a light pen, a handwriting input pad, a touch screen, and a remote control bar.
  • the input device trigger information is information directly input to the computer device after the input device is triggered.
  • the input device triggering information may specifically be a moving track of the mouse, a button frequency, a number of key presses, or the like, or a time when the button of the keyboard is pressed, a time when the button is bounced back, a key position of the pressed button, and a sequence in which the button is pressed. It can also be the coordinates touched by the touch screen and the coordinates of the touch points that are valid at the same time.
  • the second typing behavior model is a typing behavior model generated based on input device trigger information.
  • the server sends a monitoring instruction to the terminal to identify an input device for the terminal to enter the text information.
  • the server continuously sends a monitoring instruction for monitoring the input device to the terminal until the server receives the terminal.
  • Input device trigger information collected and fed back according to the monitoring command.
  • the server After the server generates the first typing model according to the collected text information and the recording interval time, and converts the text information into a typing habit model according to the typing habit feature template, the second typing behavior model is generated according to the input device trigger information.
  • the first typing model and the second typing model are combined with the typing habit model to obtain the user identity verification result.
  • the verification result can be made more accurate.
  • the input device trigger information when it is recognized that the input device is a physical keyboard, the input device trigger information includes key trigger information; generating a second typing behavior model according to the input device trigger information, including: parsing the key trigger information Obtaining a triggered key position and a corresponding key trigger interval time; generating the second typing behavior model according to the triggered key position and the corresponding key position trigger interval.
  • a physical keyboard is an input device that can input English letters, numbers, punctuation, and function commands to a computer device.
  • the key trigger information is information that the key on the physical keyboard is triggered.
  • the key trigger information may specifically be the time when the button is pressed, the time the button is bounced back, the key position of the pressed button, and the order in which the button is pressed.
  • the key trigger interval is the interval at which the keys on the keyboard are triggered.
  • the first typing behavior model is constructed according to V 0
  • the second typing behavior model is constructed according to V 1 and V 2
  • the typing habit model is compared with the first typing behavior model and the second typing behavior model.
  • the key trigger information can be acquired, and the second typing behavior model is generated according to the key trigger information, and the second typing behavior model can be used to compare with the typing habit model. , which makes user authentication more accurate.
  • the method further includes: selecting, according to the character in the text information, the triggered when the character is entered. Key position; confirm the corresponding key position trigger interval according to the selected key position; compare whether the error between the key trigger interval time and the corresponding character input interval time is within a preset range; if not, generate the When the first typing behavior model and the second typing behavior model are used, the corresponding characters and corresponding key bits are filtered.
  • Comparing the error between the key trigger interval time and the corresponding character input interval time is within a preset range, specifically, according to the order of characters, sequentially calculating the error of the key trigger interval time and the entry interval time, and detecting Whether the calculated error is within the preset range.
  • the server may generate the first typing behavior model and the second typing behavior model when the server generates the first typing behavior model and the second typing behavior model.
  • the error between the key trigger interval time and the corresponding character entry interval time is not in the preset range of key positions and character filtering, while the other time triggered key positions and entered characters, as long as the key position trigger interval time and corresponding
  • the error between the entry intervals of characters is within the preset range, and will not be filtered even if it is the same as the filtered key and character.
  • the key to trigger interval error exceeds a preset range and character input keys between the filter and the interval of the corresponding character, and in accordance with The filtered text information and the corresponding entry interval time generate a first typing behavior model, and generate a second typing behavior model according to the filtered key position and the corresponding key trigger interval time.
  • the method when it is recognized that the input device is a virtual keyboard, the method further comprises: monitoring a layout of the virtual keyboard; and when the layout change is detected, acquiring the changed layout and changing the time taken for the layout
  • the input device triggering information includes virtual key position triggering information; generating a second typing behavior model according to the input device triggering information, further comprising: parsing the virtual key position triggering information, and obtaining the triggered virtual key position and the corresponding virtual key position; Trigger interval time; calibrating the virtual key trigger interval according to the time taken to change the layout; generating the second typing behavior model according to the triggered virtual key position and the calibrated key trigger interval time.
  • a virtual keyboard is a computer readable instruction that emulates a physical keyboard.
  • the virtual keyboard can be triggered by various input devices such as a mouse, a touch screen, and a tablet.
  • the virtual keyboard can also change the layout and move the coordinate range.
  • the layout of the virtual keyboard is the distribution state of the virtual keys on the virtual keyboard.
  • the input device trigger information includes a mouse movement track; the method further includes: analyzing the mouse movement track to obtain the coordinates of the mouse at a preset time point Point, moving direction and moving speed; calculating the jitter state and moving acceleration of the mouse at the plurality of preset time points according to corresponding coordinate points, moving directions and moving speeds of the plurality of preset time points; reading the stored The jitter state and the moving acceleration are compared, and the read jitter state and the moving acceleration are compared with the calculated jitter state and the moving acceleration to obtain a comparison result; the obtained comparison result is used to correct the user identity verification result.
  • the mouse movement track may specifically be a track of the cursor movement displayed by the mouse on the computer device.
  • the coordinate point of the mouse may specifically be the coordinate point of the mouse cursor on the display screen.
  • the jitter state is specifically a state of change of the moving direction of the mouse.
  • the jitter state may be that the angle of the moving direction corresponding to two adjacent coordinate points in the moving track of the mouse exceeds a threshold value, and the image is displayed as jitter. When the angle does not exceed the threshold, the image is displayed as unjittered.
  • the server analyzes the mouse movement trajectory to obtain the coordinate point, the moving direction, and the moving speed of the mouse at a preset time point, according to corresponding coordinate points, moving directions, and moving speeds of the plurality of preset time points.
  • the jitter state of the mouse at a plurality of preset time points is calculated.
  • the server collects jitter states of multiple preset time points, obtains jitter trend information that changes with time, and reads historical jitter trend information from the database, and compares the jitter trend information with the historical jitter trend information. If the two are close, the user authentication result is not processed. If the difference between the two is large, the verification rating of the user identity in the user authentication result is lowered. When the verification rating is lower than the threshold, the user identity verification result indicates that the user identity is not the user himself.
  • the user identity verification result is corrected according to the comparison result, and the user identity can be verified by combining various input devices, so that the verification result is more accurate.
  • the method further includes: if the user identity verification result indicates that the user identity is the user, re-training the typing habit feature template according to the first typing behavior model; using the training The post-typing habit feature template updates the untrained typing habit feature template.
  • Retraining is the process of regenerating a template.
  • the retraining of the typing habit feature template may specifically be based on the first typing behavior model that trains the typing habit feature template and the first typing behavior model of the current training.
  • the server reads the typing behavior model for generating the typing habit feature template when the user identity verification result indicates that the user identity is the user himself, and selects the generation time not exceeding three months according to the frequency generated by the typing behavior model.
  • the typing behavior model combined with the typing behavior model, retrains the typing habit feature template, and writes the retrained typing habit feature template into the database to update the typing habit feature template.
  • the retraining typing habit feature template by retraining the typing habit feature template when the user identity verification result indicates that the user identity is the user himself, the retraining typing habit feature template can be made closer to the user's recent typing habit, thereby making the typing according to the typing.
  • Custom feature templates verify the user's identity more accurately.
  • a user identity verification method is further provided, which specifically includes the following steps:
  • S306. Collect text information entered by the terminal and record entry time of characters in the text information.
  • the typing habit feature template is retrained according to the first typing behavior model.
  • the user identity verification method generates a typing behavior model according to the collected text information and the recording interval time by collecting the text information entered by the terminal and the entry interval of the characters in the text information, and the typing behavior model is different for each user. Typing will not be the same.
  • the typing habit feature template extracted from the database can reflect the user's typing habits, and the text information is converted into a typing habit model through the typing habit feature template. Therefore, the typing habit model can also reflect the user's typing habits. By comparing the typing habit model with the typing behavior model, it can be compared whether the corresponding typing behavior of the typing behavior model matches the typing habits corresponding to the typing habit model.
  • the security of verification can be ensured by comparing whether typing behavior and typing habits match.
  • the text information that needs to be input is easily recognized by the user, and thus it is not easy to repeatedly verify the identification error, thereby greatly reducing the time taken for the user to verify the identity, thereby improving the efficiency of the authentication.
  • a user identity verification apparatus 400 including: an information collection module 402, a model generation module 404, a template reading module 406, and a model comparison module 408, wherein: information collection
  • the module 402 is configured to collect the text information entered by the terminal and the entry interval of the characters in the text information.
  • the model generating module 404 is configured to generate a first typing behavior model according to the text information and the entry interval time.
  • the template reading module 406 For reading the typing habit feature template, the model generating module 404 is further configured to convert the text information into a typing habit model by using the typing habit feature template; the model comparing module 408 is configured to compare the typing habit model with the first A typing behavior model that obtains user authentication results.
  • the user identity verification device 400 can generate a typing behavior model according to the collected text information and the recording interval time by collecting the text information entered by the terminal and the entry interval of the characters in the text information, and different users according to different typing habits of each user.
  • the typing behavior will not be the same.
  • the typing habit feature template extracted from the database can reflect the user's typing habits, and the text information is converted into a typing habit model through the typing habit feature template. Therefore, the typing habit model can also reflect the user's typing habits. By comparing the typing habit model with the typing behavior model, it can be compared whether the corresponding typing behavior of the typing behavior model matches the typing habits corresponding to the typing habit model.
  • the security of verification can be ensured by comparing whether typing behavior and typing habits match.
  • the text information that needs to be input is easily recognized by the user, and thus it is not easy to repeatedly verify the identification error, thereby greatly reducing the time taken for the user to verify the identity, thereby improving the efficiency of the authentication.
  • the apparatus further includes: a device identification module 410 for identifying a type of input device for inputting the text information to the terminal; and an instruction sending module 412 for identifying When the type of the input device is reached, a monitoring instruction for monitoring the input device of the type is sent to the terminal; the information collecting module 402 is further configured to receive input device trigger information collected and fed back by the terminal according to the monitoring instruction; The module 404 is further configured to generate a second typing behavior model according to the input device triggering information. The model comparison module 408 is further configured to compare the typing habit model, the first typing behavior model, and the second typing behavior model to obtain a user. The result of the authentication.
  • the apparatus further includes: an information parsing module 414, configured to parse the key trigger information, and obtain a triggered key position and a corresponding key trigger interval time; the model generation The module 404 is further configured to generate the second typing behavior model according to the triggered key position and the corresponding key trigger interval time.
  • the device further includes: a key selection module 416, configured to select a key that is triggered when the character is entered according to characters in the text information; and the time confirmation module 418 uses Confirming the corresponding key trigger interval time according to the selected key position; the time comparison module 420 is configured to compare whether the error between the key trigger interval time and the corresponding character input interval time is within a preset range; The filtering module 422 is configured to: when the error between the key trigger interval interval and the corresponding character input interval time is not within the preset range, when the first typing behavior model and the second typing behavior model are generated, The corresponding characters and corresponding key bits are filtered.
  • a key selection module 416 configured to select a key that is triggered when the character is entered according to characters in the text information
  • the time confirmation module 418 uses Confirming the corresponding key trigger interval time according to the selected key position
  • the time comparison module 420 is configured to compare whether the error between the key trigger interval time and the corresponding character input interval time is within a preset range
  • the device further includes: a layout monitoring module 424, configured to monitor a layout of the virtual keyboard; the information collection module 402 is further configured to: when the layout change is detected, Obtaining the changed layout and the time taken to change the layout; the information parsing module 414 is further configured to parse the virtual key trigger information to obtain the triggered virtual key position and the corresponding virtual key trigger interval time; the time calibration module 426 And calibrating the virtual key trigger interval according to the time taken to change the layout; the model generating module 404 is further configured to generate the first according to the triggered virtual key and the calibrated key trigger interval Two typing behavior model.
  • a layout monitoring module 424 configured to monitor a layout of the virtual keyboard
  • the information collection module 402 is further configured to: when the layout change is detected, Obtaining the changed layout and the time taken to change the layout
  • the information parsing module 414 is further configured to parse the virtual key trigger information to obtain the triggered virtual key position and the corresponding virtual key trigger interval time
  • the time calibration module 426 And calibr
  • the information parsing module 414 is further configured to analyze the mouse movement trajectory to obtain a coordinate point, a moving direction, and a moving speed of the mouse at a preset time point; corresponding to the plurality of preset time points. a coordinate point, a moving direction, and a moving speed, and calculating a jitter state and a moving acceleration of the mouse at the plurality of preset time points; the model comparing module 408 is further configured to read the stored jitter state and the moving acceleration, and The read jitter state and the moving acceleration are compared with the calculated jitter state and the moving acceleration to obtain a comparison result; the obtained comparison result is used to correct the user identity verification result.
  • the apparatus further includes: a retraining module 428, configured to: if the user identity verification result indicates that the user identity is the user, the typing according to the first typing behavior model The habit feature template is retrained; the untrained typing habit feature template is updated using the trained typing habit feature template.
  • a retraining module 428 configured to: if the user identity verification result indicates that the user identity is the user, the typing according to the first typing behavior model The habit feature template is retrained; the untrained typing habit feature template is updated using the trained typing habit feature template.
  • Each of the above-described user authentication devices may be implemented in whole or in part by software, hardware, and combinations thereof.
  • Each of the above modules may be embedded in or independent of the processor in the computer device, or may be stored in a memory in the computer device in a software form, so that the processor invokes the operations corresponding to the above modules.
  • a computer device which may be a server, and its internal structure diagram may be as shown in FIG.
  • the computer device includes a processor, memory, network interface, and database connected by a system bus.
  • the processor of the computer device is used to provide computing and control capabilities.
  • the memory of the computer device includes a non-volatile storage medium, an internal memory.
  • the non-volatile storage medium stores an operating system, computer readable instructions, and a database.
  • the internal memory provides an environment for operation of an operating system and computer readable instructions in a non-volatile storage medium.
  • the database of the computer device is used to store data such as typing habit feature templates.
  • the network interface of the computer device is used to communicate with an external terminal via a network connection.
  • the computer readable instructions are executed by a processor to implement a user authentication method.
  • FIG. 10 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation of the computer device to which the solution of the present application is applied.
  • the specific computer device may It includes more or fewer components than those shown in the figures, or some components are combined, or have different component arrangements.
  • a computer apparatus comprising a memory, a processor, and computer readable instructions stored on the memory and executable on the processor, the processor implementing any one of the present applications when executing the computer readable instructions The steps of the user authentication method provided in the embodiment.
  • the computer device can generate a typing behavior model according to the collected text information and the recording interval time by collecting the text information entered by the terminal and the entry interval of the characters in the text information, and the typing behavior of different users due to different typing habits of each user It will not be the same.
  • the typing habit feature template extracted from the database can reflect the user's typing habits, and the text information is converted into a typing habit model through the typing habit feature template. Therefore, the typing habit model can also reflect the user's typing habits. By comparing the typing habit model with the typing behavior model, it can be compared whether the corresponding typing behavior of the typing behavior model matches the typing habits corresponding to the typing habit model.
  • the security of verification can be ensured by comparing whether typing behavior and typing habits match.
  • the text information that needs to be input is easily recognized by the user, and thus it is not easy to repeatedly verify the identification error, thereby greatly reducing the time taken for the user to verify the identity, thereby improving the efficiency of the authentication.
  • non-volatile storage media having computer readable instructions that, when executed by one or more processors, cause one or more processors to implement the present The steps of applying the user authentication method provided in any of the embodiments.
  • the computer readable storage medium can generate a typing behavior model according to the collected text information and the recording interval time by collecting the text information entered by the terminal and the entry interval of the characters in the text information, and different users according to different typing habits of each user The typing behavior will not be the same.
  • the typing habit feature template extracted from the database can reflect the user's typing habits, and the text information is converted into a typing habit model through the typing habit feature template. Therefore, the typing habit model can also reflect the user's typing habits. By comparing the typing habit model with the typing behavior model, it can be compared whether the corresponding typing behavior of the typing behavior model matches the typing habits corresponding to the typing habit model.
  • the security of verification can be ensured by comparing whether typing behavior and typing habits match.
  • the text information that needs to be input is easily recognized by the user, and thus it is not easy to repeatedly verify the identification error, thereby greatly reducing the time taken for the user to verify the identity, thereby improving the efficiency of the authentication.
  • Non-volatile memory can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory can include random access memory (RAM) or external cache memory.
  • RAM is available in a variety of formats, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronization chain.
  • SRAM static RAM
  • DRAM dynamic RAM
  • SDRAM synchronous DRAM
  • DDRSDRAM double data rate SDRAM
  • ESDRAM enhanced SDRAM
  • Synchlink DRAM SLDRAM
  • Memory Bus Radbus
  • RDRAM Direct RAM
  • DRAM Direct Memory Bus Dynamic RAM
  • RDRAM Memory Bus Dynamic RAM

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • Social Psychology (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种用户身份验证方法,包括:采集终端录入的文本信息和该文本信息中字符的录入间隔时间;根据该文本信息和该录入间隔时间生成第一打字行为模型;读取打字习惯特征模板;通过该打字习惯特征模板将该文本信息转换成打字习惯模型;对比该打字习惯模型和该第一打字行为模型,得到用户身份验证结果。

Description

用户身份验证方法、装置、计算机设备和存储介质
相关申请的交叉引用
本申请要求于2018年02月08日提交中国专利局,申请号为2018101283701,申请名称为“用户身份验证方法、装置、计算机设备和存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及一种用户身份验证方法、装置、计算机设备和存储介质。
背景技术
随着互联网技术的发展,信息安全越来越重要,为了保护信息的安全,通常用户会给自己的账号设置密码。而恶意用户则主要是通过撞库等暴力破解方式来破解密码,由于这类暴力破解方式是通过计算机可读指令重复尝试密码。因此在设置密码的同时,设置一个验证码可以有效地防范这一类暴力破解方式。
然而,发明人意识到,传统的验证码大多都是图形验证码,随着智能识图技术的发展,简单的图形验证码很容易被识图程序识别,使得简单的图形验证码无法防范计算机可读指令。而复杂的图形验证码则会使得用户也难以识别,用户在使用复杂的图形验证码验证时也常常会出错,使得验证过程花费的时间过长,效率很低。
发明内容
根据本申请公开的各种实施例,提供一种能够提高验证效率的用户身份验证方法、装置、计算机设备和存储介质。
一种用户身份验证方法,包括:
采集终端录入的文本信息和该文本信息中字符的录入间隔时间;
根据该文本信息和该录入间隔时间生成第一打字行为模型;
读取打字习惯特征模板;
通过该打字习惯特征模板将该文本信息转换成打字习惯模型;及
对比该打字习惯模型和该第一打字行为模型,得到用户身份验证结果。
一种用户身份验证装置,包括:
信息采集模块,用于采集终端录入的文本信息和该文本信息中字符的录入间隔时间;
模型生成模块,用于根据该文本信息和该录入间隔时间生成第一打字行为模型;
模板读取模块,用于读取打字习惯特征模板;
该模型生成模块,还用于通过该打字习惯特征模板将该文本信息转换成打字习惯模 型;及
模型对比模块,用于对比该打字习惯模型和该第一打字行为模型,得到用户身份验证结果。
一种计算机设备,包括存储器和一个或多个处理器,所述存储器中储存有计算机可读指令,所述计算机可读指令被所述处理器执行时,使得所述一个或多个处理器执行以下步骤:
采集终端录入的文本信息和该文本信息中字符的录入间隔时间;
根据该文本信息和该录入间隔时间生成第一打字行为模型;
读取打字习惯特征模板;
通过该打字习惯特征模板将该文本信息转换成打字习惯模型;及
对比该打字习惯模型和该第一打字行为模型,得到用户身份验证结果。
一个或多个存储有计算机可读指令的非易失性存储介质,计算机可读指令被一个或多个处理器执行时,使得一个或多个处理器执行以下步骤:
采集终端录入的文本信息和该文本信息中字符的录入间隔时间;
根据该文本信息和该录入间隔时间生成第一打字行为模型;
读取打字习惯特征模板;
通过该打字习惯特征模板将该文本信息转换成打字习惯模型;及
对比该打字习惯模型和该第一打字行为模型,得到用户身份验证结果。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其它的附图。
图1为根据一个或多个实施例中用户身份验证方法的应用场景图;
图2为根据一个或多个实施例中用户身份验证方法的流程示意图;
图3为另一个实施例中用户身份验证方法的流程示意图;
图4为根据一个或多个实施例中用户身份验证装置的框图;
图5为另一个实施例中用户身份验证装置的框图;
图6为根据一个或多个实施例中用户身份验证装置的框图;
图7为另一个实施例中用户身份验证装置的框图;
图8为根据一个或多个实施例中用户身份验证装置的框图;
图9为另一个实施例中用户身份验证装置的框图;
图10为根据一个或多个实施例中计算机设备的框图。
具体实施方式
为了使本申请的技术方案及优点更加清楚明白,以下结合附图及实施例,对本申请进行进一步详细说明。应当理解,此处描述的具体实施例仅仅用以解释本申请,并不用于限定本申请。
本申请提供的用户身份验证方法,可以应用于如图1所示的应用环境中。终端110通过网络与服务器120通过网络进行通信。服务器120采集终端110录入的文本信息和文本信息中字符的录入间隔时间,根据文本信息和录入间隔时间生成第一打字行为模型。服务器120还通过读取打字习惯特征模板,使用打字习惯特征模板将文本信息转换成打字习惯模型,并对比打字习惯模型和第一打字行为模型,得到用户身份验证结果。
终端110可以但不限于是各种个人计算机、笔记本电脑、智能手机、平板电脑和便携式可穿戴设备,服务器120可以用独立的服务器或者是多个服务器组成的服务器集群来实现。
在其中一个实施例中,如图2所示,提供了一种用户身份验证方法,以该方法应用于图1中的服务器为例进行说明,包括以下步骤:
S202,采集终端录入的文本信息和该文本信息中字符的录入间隔时间;
终端可以是图1中的终端。文本信息是记载文字信息的文本文件。文字信息具体可以是通过不同字符以不同排列组合表达的内容。终端录入文本信息,具体可以是通过输入装置在服务器提供的页面上的文本输入框中录入文字信息,并将文字信息保存在文本信息中。字符是一个单位的字形、类字形单位或符号的基本信息。字符具体可以是字母、数字、文字和符号等。录入间隔时间是录入两个字符的时间点之间的时间段。
在其中一个实施例中,服务器通过发送到终端的网页上的JavaScript脚本监测该网页上的文本框。当服务器监测到终端在文本框中录入字符时,采集终端录入的字符和终端录入字符的时间点。当终端录入字符完毕后,将所有的字符保存在文本信息中,并根据终端录入字符的时间点计算得到文本信息中相邻两个字符的录入间隔时间,以及文本信息的录入时间。
在其中一个实施例中,一种用户身份验证方法还应用在终端上,终端在触发录入文本信息的指令后,实时监测与终端连接的输入装置,采集输入装置输入的字符和输入装置输入字符的时间点,并根据终端上存储的输入法程序将输入装置输入的字符转换成录入文本信息中的字符。在文本信息中的字符录入完毕后,根据输入装置输入的字符和输入装置输入字符的时间点生成文本信息中字符的录入间隔时间。
S204,根据该文本信息和该录入间隔时间生成第一打字行为模型。
打字行为模型具体可以是阐述文本信息相应打字行为的工具。打字行为模型具体可以是数学公式,也可以是图表。打字行为是通过输入装置输入文本信息的行为。打字行为具 体可以是敲击实体键盘上按键的行为,例如用户击键行为,也可以是通过触摸触控面板或操作鼠标等,以触发虚拟键盘上的虚拟按键的行为。
在其中一个实施例中,服务器根据文本信息中字符的排列顺序,依次将排列录入间隔时间,并根据依次排列的录入间隔时间生成录入间隔时间序列。服务器根据录入间隔时间序列生成第一打字行为模型。
在其中一个实施例中,一种用户身份验证方法还应用在终端上,终端根据输入装置输入的字符和输入装置输入字符的时间点生成输入间隔时间序列,并根据文本信息和相应的录入间隔时间生成录入间隔时间序列。终端根据输入间隔时间序列和录入间隔时间序列生成第一打字行为模型。
S206,读取打字习惯特征模板。
打字习惯特征模板是记录打字习惯的特征的模板。打字习惯是打字行为的特征集合。具体的,假设存在根据一个用户所有的打字行为生成的打字习惯,则根据该打字习惯录入的文本信息都与用户本人录入该文本信息的打字行为相匹配。
在其中一个实施例中,服务器从终端获取用户身份标识,根据用户身份标识读取相应的打字习惯特征模板。服务器还检测读取的打字习惯特征模板的再训练次数和最后一次再训练的时间,当检测到再训练次数和/或最后一次再训练的时间超过预设阈值,则多次向终端发送录入文本信息的指令。
S208,通过该打字习惯特征模板将该文本信息转换成打字习惯模型。
打字习惯模型是根据打字习惯特征模板模拟出的打字行为模型。具体的,通过打字习惯特征模板将文本信息转换成打字习惯模型,可以是在打字习惯特征模板中选取与文本信息中字符的排列顺序匹配的特征,例如按照文本信息中字符的排列顺序排列的录入间隔时间。
在其中一个实施例中,服务器根据文本信息中的字符的排列顺序,在打字习惯特征模板选取相应的间隔时间,并按照字符的排列顺序将相应的间隔时间排列,根据排列后的间隔时间生成打字习惯模型。
S210,对比该打字习惯模型和该第一打字行为模型,得到用户身份验证结果。
本实施例中,通过采集终端录入的文本信息和文本信息中字符的录入间隔时间,根据采集的文本信息和录入间隔时间可以生成打字行为模型,由于每个用户打字习惯的不同,不同用户的打字行为也就不会相同。从数据库中提取的打字习惯特征模板可以反映用户的打字习惯,通过打字习惯特征模板将文本信息转换成打字习惯模型,因此打字习惯模型也可以反映用户的打字习惯。通过对比打字习惯模型和打字行为模型,可以比较出打字行为模型相应的打字行为是否与打字习惯模型相应的打字习惯相匹配。由于打字习惯难以在未经过长期学习和训练的情况下被计算机可读指令模拟,因此通过对比打字行为和打字习惯是否匹配可以保证验证的安全性。而且需要输入的文本信息很容易被用户识别,因而也不容易发生因识别错误而反复验证的情况,从而极大地减少了用户验证身份所花费的时间, 进而提高了身份验证的效率。
在其中一个实施例中,采集终端录入的文本信息之后,该方法还包括:识别用于向该终端录入该文本信息的输入装置的类型;当识别到该输入装置的类型时,向该终端发送监测该类型的输入装置的监测指令;接收该终端根据该监测指令采集并反馈的输入装置触发信息;步骤S208之后,该方法还包括:根据该输入装置触发信息生成第二打字行为模型;对比该打字习惯模型、该第一打字行为模型和该第二打字行为模型,得到用户身份验证结果。
输入装置是向计算机设备输入数据和信息的装置。输入装置具体可以是键盘、鼠标、光笔、手写输入板、触控屏和遥控杆等中的至少一种。输入装置触发信息是输入装置被触发后向计算机设备直接输入的信息。输入装置触发信息具体可以是鼠标的移动轨迹、按键频率和按键次数等,也可以是键盘的按键被按的时间、按键弹回的时间、被按按键的键位和按键被按的顺序等,还可以是触控屏被触摸的坐标和同一时间有效的触控点的坐标等。第二打字行为模型是根据输入装置触发信息生成的打字行为模型。
在其中一个实施例中,服务器向终端发送监测指令,以识别终端录入文本信息的输入装置,当识别出该输入装置时,服务器持续向终端发送监测该输入装置的监测指令,直至服务器接收到终端根据该监测指令采集并反馈的输入装置触发信息。当服务器根据采集的文本信息和录入间隔时间生成第一打字模型后,以及根据打字习惯特征模板将文本信息转换成打字习惯模型后,根据输入装置触发信息生成第二打字行为模型。结合第一打字模型和第二打字模型对比打字习惯模型,得到用户身份验证结果。
本实施例中,通过识别终端录入文本信息的输入装置,可以更准确地判断终端反馈的输入装置触发信息是否准确。而且通过根据输入装置触发信息生成第二打字行为模型,对比打字习惯模型、第一打字行为模型和第二打字行为模型,可以使得验证结果更加准确。
在其中一个实施例中,当识别到该输入装置是实体键盘时,该输入装置触发信息包括键位触发信息;根据该输入装置触发信息生成第二打字行为模型,包括:解析该键位触发信息,得到被触发的键位和相应的键位触发间隔时间;根据该被触发的键位和相应的键位触发间隔时间生成该第二打字行为模型。
实体键盘是一种可以向计算机设备输入英文字母、数字、标点符号和功能指令等的输入装置。键位触发信息是实体键盘上键位被触发的信息。键位触发信息具体可以是按键被按的时间、按键弹回的时间、被按按键的键位和按键被按的顺序等。键位触发间隔时间是键盘上按键被触发的间隔时间。
在其中一个实施例中,实体键盘的按键被按下的时间点是P,按键回弹的时间点是R,则按照按键被触发的顺序,采集到按键序列V 0={P 1,R 1,P 2,R 2,···,P n,R n},其中n为大于2的整数。则根据V 0可以获取按键持续时间序列V 1,按键持续时间为R-P,定义PR i=R i-P i(i=1,···,n),则V 1={PR 1,PR 2,···,PR i}。并可以根据V 0可以获取按键间隔时间序列V 2,按键持续时间为P-R,定义RP i=P i+1-R i(i=1,···,n-1),则V 2={RP 1, RP 2,···,RP i}。根据V 0构建第一打字行为模型,根据V 1和V 2构建第二打字行为模型,并结合第一打字行为模型和第二打字行为模型对比打字习惯模型。
本实施例中,通过在识别到输入装置是实体键盘时,可以获取到键位触发信息,并根据键位触发信息生成第二打字行为模型,第二打字行为模型可以用于和打字习惯模型对比,从而使得用户身份验证更加精准。
在其中一个实施例中,解析该键位触发信息,得到被触发的键位和相应的键位触发间隔时间之后,该方法还包括:根据该文本信息中的字符选取录入该字符时被触发的键位;根据选取的键位确认相应的键位触发间隔时间;对比该键位触发间隔时间和相应的字符的录入间隔时间之间的误差是否在预设范围内;若否,则在生成该第一打字行为模型和该第二打字行为模型时,将相应的字符和相应的键位过滤。
对比键位触发间隔时间和相应的字符的录入间隔时间之间的误差是否在预设范围内,具体可以是按照字符的排列顺序,依次计算键位触发间隔时间和录入间隔时间的误差,并检测计算出的误差是否在预设范围内。具体地,在生成该第一打字行为模型和该第二打字行为模型时,将相应的字符和相应的键位过滤,可以是服务器在生成第一打字行为模型和第二打字行为模型时,将键位触发间隔时间和相应的字符的录入间隔时间之间的误差不在预设范围内的键位和字符过滤,而其它时间触发的键位和录入的字符,只要键位触发间隔时间和相应的字符的录入间隔时间之间的误差在预设范围内,就算与被过滤键位和字符相同,也不会被过滤。
在其中一个实施例中,服务器根据文本信息中的字符选取录入该字符时被触发的键位,定义实体键盘的按键被按下的时间点是P,则采集到键位触发间隔时间序列V 3={P 2-P 1,P 3-P 2,···,P n-P n-1},其中n为大于3的整数。并且使用采集到的文本信息中字符的录入间隔时间Vx与V 3进行对比,将键位触发间隔时间与相应的字符的录入间隔时间之间误差超过预设范围的字符和键位过滤,并根据过滤后的文本信息和相应的录入间隔时间生成第一打字行为模型,根据过滤后的键位和相应的键位触发间隔时间生成第二打字行为模型。
本实施例中,通过对比键位触发间隔时间和相应的字符的录入间隔时间之间的误差是否在预设范围内,可以在生成第一打字行为模型和第二打字行为模型时,将相应的字符和键位过滤,从而可以避免由于环境因素导致获取到的间隔时间不准确,导致生成的打字行为模型误差较大。
在其中一个实施例中,当识别到该输入装置是虚拟键盘时,该方法还包括:监测该虚拟键盘的布局;当监测到该布局改变时,则获取改变后的布局和改变布局花费的时间;该输入装置触发信息包括虚拟键位触发信息;根据该输入装置触发信息生成第二打字行为模型,还包括:解析该虚拟键位触发信息,得到被触发的虚拟键位和相应的虚拟键位触发间隔时间;根据该改变布局花费的时间对该虚拟键位触发间隔时间进行校准;根据该被触发的虚拟键位和校准后的键位触发间隔时间生成该第二打字行为模型。虚拟键盘是模拟实体键盘的计算机可读指令。虚拟键盘可以通过被鼠标、触控屏和手写板等多种输入装置触发, 虚拟键盘还可以更改布局和移动坐标范围。虚拟键盘的布局是虚拟键盘上虚拟键位的分布状态。
本实施例中,通过监测虚拟键盘的布局和改变布局花费的时间,可以避免由于改变布局时花费的时间干扰采集到的虚拟键位触发间隔时间,从而可以提高用户身份验证的精准度。
在其中一个实施例中,当识别到该输入装置还包括鼠标时,该输入装置触发信息包括鼠标移动轨迹;该方法还包括:分析该鼠标移动轨迹,得到在预设时间点时该鼠标的坐标点、移动方向和移动速度;根据多个预设时间点相应的坐标点、移动方向和移动速度,计算得出该鼠标在该多个预设时间点的抖动状态和移动加速度;读取存储的抖动状态和移动加速度,并将读取的抖动状态和移动加速度和计算出的抖动状态和移动加速度进行对比,得到对比结果;使用得到的对比结果对该用户身份验证结果进行修正。
鼠标移动轨迹具体可以是鼠标在计算机设备上显示的光标移动的轨迹。鼠标的坐标点具体可以是鼠标的光标在显示屏上的坐标点。抖动状态具体是鼠标的移动方向的变化状态。抖动状态具体可以是鼠标的移动轨迹中两个相邻坐标点对应的移动方向的夹角超过阈值时显示为抖动,夹角未超过阈值时显示为未抖动。
在其中一个实施例中,服务器分析鼠标移动轨迹,得到在预设时间点时该鼠标的坐标点、移动方向和移动速度,根据多个预设时间点相应的坐标点、移动方向和移动速度,计算得出该鼠标在多个预设时间点的抖动状态。服务器统计多个预设时间点的抖动状态,得到随着时间变化的抖动趋势信息,并从数据库中读取历史抖动趋势信息,对比该抖动趋势信息和历史抖动趋势信息。若两者较为接近,则不对用户身份验证结果进行处理。若两者区别较大,则降低用户身份验证结果中对用户身份的验证评级,在验证评级低于阈值时,用户身份验证结果表示用户身份不是用户本人。
本实施例中,通过根据鼠标移动轨迹生成抖动状态和移动加速度,并对比抖动状态和移动加速度,根据对比结果对用户身份验证结果进行修正,可以结合多种输入装置验证用户身份,使得验证结果更加准确。
在其中一个实施例中,步骤S210之后,该方法还包括:若该用户身份验证结果表示用户身份是用户本人时,则根据该第一打字行为模型对该打字习惯特征模板进行再训练;使用训练后的打字习惯特征模板更新未训练的打字习惯特征模板。
再训练是重新生成模板的过程。对打字习惯特征模板进行再训练具体可以是根据训练出打字习惯特征模板的第一打字行为模型和本次的第一打字行为模型进行训练。
在其中一个实施例中,服务器在用户身份验证结果表示用户身份是用户本人时,读取生成打字习惯特征模板的打字行为模型,按照打字行为模型生成的频率,选取生成时间不超过三个月的打字行为模型,并结合本次的打字行为模型对打字习惯特征模板进行再训练,将再训练后的打字习惯特征模板写入数据库中,以更新打字习惯特征模板。
本实施例中,通过在用户身份验证结果表示用户身份是用户本人时,对打字习惯特 征模板进行再训练,可以使得再训练的打字习惯特征模板与用户最近的打字习惯更加接近,从而使得根据打字习惯特征模板验证用户身份的准确性更高。
如图3所示,在其中一个实施例中,还提供了一种用户身份验证方法,具体包括以下的步骤:
S302,识别终端录入文本信息的输入装置。
S304,当识别到该输入装置是实体键盘和鼠标时,向终端发送监测实体键盘和鼠标的监测指令。
S306,采集终端录入的文本信息和文本信息中字符的录入间隔时间。
S308,接收终端根据监测指令采集并反馈的键位触发信息和鼠标移动轨迹。
S310,根据文本信息和录入间隔时间生成第一打字行为模型。
S312,解析键位触发信息,得到被触发的键位和相应的键位触发间隔时间。
S314,根据被触发的键位和相应的键位触发间隔时间生成第二打字行为模型。
S316,读取打字习惯特征模板。
S318,通过打字习惯特征模板将文本信息转换成打字习惯模型。
S320,结合第一打字行为模型和第二打字行为模型对比打字习惯模型,得到用户身份验证结果。
S322,分析鼠标移动轨迹,得到鼠标在多个预设时间点的抖动状态。
S324,读取存储的抖动状态,并将读取的抖动状态和得到的抖动状态进行对比,得到对比结果。
S326,使用得到的对比结果对用户身份验证结果进行修正。
S328,当修正后的用户身份验证结果表示用户身份是用户本人时,则根据第一打字行为模型对打字习惯特征模板进行再训练。
S330,使用训练后的打字习惯特征模板更新未训练的打字习惯特征模板。
上述用户身份验证方法,通过采集终端录入的文本信息和文本信息中字符的录入间隔时间,根据采集的文本信息和录入间隔时间可以生成打字行为模型,由于每个用户打字习惯的不同,不同用户的打字行为也就不会相同。从数据库中提取的打字习惯特征模板可以反映用户的打字习惯,通过打字习惯特征模板将文本信息转换成打字习惯模型,因此打字习惯模型也可以反映用户的打字习惯。通过对比打字习惯模型和打字行为模型,可以比较出打字行为模型相应的打字行为是否与打字习惯模型相应的打字习惯相匹配。由于打字习惯难以在未经过长期学习和训练的情况下被计算机可读指令模拟,因此通过对比打字行为和打字习惯是否匹配可以保证验证的安全性。而且需要输入的文本信息很容易被用户识别,因而也不容易发生因识别错误而反复验证的情况,从而极大地减少了用户验证身份所花费的时间,进而提高了身份验证的效率。
应该理解的是,虽然图2和3的流程图中的各个步骤按照箭头的指示依次显示,但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明,这些步 骤的执行并没有严格的顺序限制,这些步骤可以以其它的顺序执行。而且,图2和3中的至少一部分步骤可以包括多个子步骤或者多个阶段,这些子步骤或者阶段并不必然是在同一时刻执行完成,而是可以在不同的时刻执行,这些子步骤或者阶段的执行顺序也不必然是依次进行,而是可以与其它步骤或者其它步骤的子步骤或者阶段的至少一部分轮流或者交替地执行。
在其中一个实施例中,如图4所示,提供了一种用户身份验证装置400,包括:信息采集模块402、模型生成模块404、模板读取模块406和模型对比模块408,其中:信息采集模块402,用于采集终端录入的文本信息和该文本信息中字符的录入间隔时间;模型生成模块404,用于根据该文本信息和该录入间隔时间生成第一打字行为模型;模板读取模块406,用于读取打字习惯特征模板;该模型生成模块404,还用于通过该打字习惯特征模板将该文本信息转换成打字习惯模型;模型对比模块408,用于对比该打字习惯模型和该第一打字行为模型,得到用户身份验证结果。
上述用户身份验证装置400,通过采集终端录入的文本信息和文本信息中字符的录入间隔时间,根据采集的文本信息和录入间隔时间可以生成打字行为模型,由于每个用户打字习惯的不同,不同用户的打字行为也就不会相同。从数据库中提取的打字习惯特征模板可以反映用户的打字习惯,通过打字习惯特征模板将文本信息转换成打字习惯模型,因此打字习惯模型也可以反映用户的打字习惯。通过对比打字习惯模型和打字行为模型,可以比较出打字行为模型相应的打字行为是否与打字习惯模型相应的打字习惯相匹配。由于打字习惯难以在未经过长期学习和训练的情况下被计算机可读指令模拟,因此通过对比打字行为和打字习惯是否匹配可以保证验证的安全性。而且需要输入的文本信息很容易被用户识别,因而也不容易发生因识别错误而反复验证的情况,从而极大地减少了用户验证身份所花费的时间,进而提高了身份验证的效率。
如图5所示,在其中一个实施例中,该装置还包括:装置识别模块410,用于识别用于向该终端录入该文本信息的输入装置的类型;指令发送模块412,用于当识别到该输入装置的类型时,向该终端发送监测该类型的输入装置的监测指令;该信息采集模块402,还用于接收该终端根据该监测指令采集并反馈的输入装置触发信息;该模型生成模块404,还用于根据该输入装置触发信息生成第二打字行为模型;该模型对比模块408,还用于对比该打字习惯模型、该第一打字行为模型和该第二打字行为模型,得到用户身份验证结果。
如图6所示,在其中一个实施例中,该装置还包括:信息解析模块414,用于解析该键位触发信息,得到被触发的键位和相应的键位触发间隔时间;该模型生成模块404,还用于根据该被触发的键位和相应的键位触发间隔时间生成该第二打字行为模型。
如图7所示,在其中一个实施例中,该装置还包括:键位选取模块416,用于根据该文本信息中的字符选取录入该字符时被触发的键位;时间确认模块418,用于根据选取的键位确认相应的键位触发间隔时间;时间对比模块420,用于对比该键位触发间隔时间和相应的字符的录入间隔时间之间的误差是否在预设范围内;键位过滤模块422,用于在键 位触发间隔时间和相应的字符的录入间隔时间之间的误差不在预设范围内时,则在生成该第一打字行为模型和该第二打字行为模型时,将相应的字符和相应的键位过滤。
如图8所示,在其中一个实施例中,该装置还包括:布局监测模块424,用于监测该虚拟键盘的布局;该信息采集模块402,还用于当监测到该布局改变时,则获取改变后的布局和改变布局花费的时间;该信息解析模块414,还用于解析该虚拟键位触发信息,得到被触发的虚拟键位和相应的虚拟键位触发间隔时间;时间校准模块426,用于根据该改变布局花费的时间对该虚拟键位触发间隔时间进行校准;该模型生成模块404,还用于根据该被触发的虚拟键位和校准后的键位触发间隔时间生成该第二打字行为模型。
在其中一个实施例中,该信息解析模块414,还用于分析该鼠标移动轨迹,得到在预设时间点时该鼠标的坐标点、移动方向和移动速度;根据多个预设时间点相应的坐标点、移动方向和移动速度,计算得出该鼠标在该多个预设时间点的抖动状态和移动加速度;该模型对比模块408,还用于读取存储的抖动状态和移动加速度,并将读取的抖动状态和移动加速度和计算出的抖动状态和移动加速度进行对比,得到对比结果;使用得到的对比结果对该用户身份验证结果进行修正。
如图9所示,在其中一个实施例中,该装置还包括:再训练模块428,用于若该用户身份验证结果表示用户身份是用户本人时,则根据该第一打字行为模型对该打字习惯特征模板进行再训练;使用训练后的打字习惯特征模板更新未训练的打字习惯特征模板。
关于用户身份验证装置的具体限定可以参见上文中对于用户身份验证方法的限定,在此不再赘述。上述用户身份验证装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中,也可以以软件形式存储于计算机设备中的存储器中,以便于处理器调用执行以上各个模块对应的操作。
在其中一个实施例中,提供了一种计算机设备,该计算机设备可以是服务器,其内部结构图可以如图10所示。该计算机设备包括通过系统总线连接的处理器、存储器、网络接口和数据库。该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统、计算机可读指令和数据库。该内存储器为非易失性存储介质中的操作系统和计算机可读指令的运行提供环境。该计算机设备的数据库用于存储打字习惯特征模板等数据。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机可读指令被处理器执行时以实现一种用户身份验证方法。
本领域技术人员可以理解,图10中示出的结构,仅仅是与本申请方案相关的部分结构的框图,并不构成对本申请方案所应用于其上的计算机设备的限定,具体的计算机设备可以包括比图中所示更多或更少的部件,或者组合某些部件,或者具有不同的部件布置。
在其中一个实施例中,提供了一种计算机设备,包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机可读指令,处理器执行计算机可读指令时实现本申请任意 一个实施例中提供的用户身份验证方法的步骤。
上述计算机设备,通过采集终端录入的文本信息和文本信息中字符的录入间隔时间,根据采集的文本信息和录入间隔时间可以生成打字行为模型,由于每个用户打字习惯的不同,不同用户的打字行为也就不会相同。从数据库中提取的打字习惯特征模板可以反映用户的打字习惯,通过打字习惯特征模板将文本信息转换成打字习惯模型,因此打字习惯模型也可以反映用户的打字习惯。通过对比打字习惯模型和打字行为模型,可以比较出打字行为模型相应的打字行为是否与打字习惯模型相应的打字习惯相匹配。由于打字习惯难以在未经过长期学习和训练的情况下被计算机可读指令模拟,因此通过对比打字行为和打字习惯是否匹配可以保证验证的安全性。而且需要输入的文本信息很容易被用户识别,因而也不容易发生因识别错误而反复验证的情况,从而极大地减少了用户验证身份所花费的时间,进而提高了身份验证的效率。
在其中一个实施例中,提供了一个或多个存储有计算机可读指令的非易失性存储介质,计算机可读指令被一个或多个处理器执行时,使得一个或多个处理器实现本申请任意一个实施例中提供的用户身份验证方法的步骤。
上述计算机可读存储介质,通过采集终端录入的文本信息和文本信息中字符的录入间隔时间,根据采集的文本信息和录入间隔时间可以生成打字行为模型,由于每个用户打字习惯的不同,不同用户的打字行为也就不会相同。从数据库中提取的打字习惯特征模板可以反映用户的打字习惯,通过打字习惯特征模板将文本信息转换成打字习惯模型,因此打字习惯模型也可以反映用户的打字习惯。通过对比打字习惯模型和打字行为模型,可以比较出打字行为模型相应的打字行为是否与打字习惯模型相应的打字习惯相匹配。由于打字习惯难以在未经过长期学习和训练的情况下被计算机可读指令模拟,因此通过对比打字行为和打字习惯是否匹配可以保证验证的安全性。而且需要输入的文本信息很容易被用户识别,因而也不容易发生因识别错误而反复验证的情况,从而极大地减少了用户验证身份所花费的时间,进而提高了身份验证的效率。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机可读指令来指令相关的硬件来完成,所述的计算机可读指令可存储于一非易失性计算机可读取存储介质中,该计算机可读指令在执行时,可包括如上述各方法的实施例的流程。本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用,均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限,RAM以多种形式可得,诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
以上实施例的各技术特征可以进行任意的组合,为使描述简洁,未对上述实施例中的各个技术特征所有可能的组合都进行描述,然而,只要这些技术特征的组合不存在矛盾,都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式,其描述较为具体和详细,但并不能因此而理解为对发明专利范围的限制。应当指出的是,对于本领域的普通技术人员来说,在不脱离本申请构思的前提下,还可以做出若干变形和改进,这些都属于本申请的保护范围。因此,本申请专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种用户身份验证方法,包括:
    采集终端录入的文本信息和所述文本信息中字符的录入间隔时间;
    根据所述文本信息和所述录入间隔时间生成第一打字行为模型;
    读取打字习惯特征模板;
    通过所述打字习惯特征模板将所述文本信息转换成打字习惯模型;及
    对比所述打字习惯模型和所述第一打字行为模型,得到用户身份验证结果。
  2. 根据权利要求1所述的方法,其特征在于,在所述采集终端录入的文本信息之后,所述方法还包括:
    识别用于向所述终端录入所述文本信息的输入装置的类型;
    当识别到所述输入装置的类型时,向所述终端发送监测所述类型的输入装置的监测指令;及
    接收所述终端根据所述监测指令采集并反馈的输入装置触发信息;
    在所述通过所述打字习惯特征模板将所述文本信息转换成打字习惯模型之后,所述方法还包括:
    根据所述输入装置触发信息生成第二打字行为模型;及
    对比所述打字习惯模型、所述第一打字行为模型和所述第二打字行为模型,得到用户身份验证结果。
  3. 根据权利要求2所述的方法,其特征在于,当识别到所述输入装置是实体键盘时,所述输入装置触发信息包括键位触发信息;所述根据所述输入装置触发信息生成第二打字行为模型,包括:
    解析所述键位触发信息,得到被触发的键位和相应的键位触发间隔时间;及
    根据所述被触发的键位和相应的键位触发间隔时间生成所述第二打字行为模型。
  4. 根据权利要求3所述的方法,其特征在于,在所述解析所述键位触发信息,得到被触发的键位和相应的键位触发间隔时间之后,所述方法还包括:
    根据所述文本信息中的字符选取录入所述字符时被触发的键位;
    根据选取的键位确认相应的键位触发间隔时间;及
    对比所述键位触发间隔时间和相应的字符的录入间隔时间之间的误差是否在预设范围内;若否,则
    在生成所述第一打字行为模型和所述第二打字行为模型时,将相应的字符和相应的键位过滤。
  5. 根据权利要求2所述的方法,其特征在于,当识别到所述输入装置是虚拟键盘时,所述方法还包括:
    监测所述虚拟键盘的布局;及
    当监测到所述布局改变时,则获取改变后的布局和改变布局花费的时间;
    所述输入装置触发信息包括虚拟键位触发信息;所述根据所述输入装置触发信息生成第 二打字行为模型,还包括:
    解析所述虚拟键位触发信息,得到被触发的虚拟键位和相应的虚拟键位触发间隔时间;
    根据所述改变布局花费的时间对所述虚拟键位触发间隔时间进行校准;及
    根据所述被触发的虚拟键位和校准后的键位触发间隔时间生成所述第二打字行为模型。
  6. 根据权利要求2所述的方法,其特征在于,当识别到所述输入装置还包括鼠标时,所述输入装置触发信息包括鼠标移动轨迹;所述方法还包括:
    分析所述鼠标移动轨迹,得到在预设时间点时所述鼠标的坐标点、移动方向和移动速度;
    根据多个预设时间点相应的坐标点、移动方向和移动速度,计算得出所述鼠标在所述多个预设时间点的抖动状态和移动加速度;
    读取存储的抖动状态和移动加速度,并将读取的抖动状态和移动加速度和计算出的抖动状态和移动加速度进行对比,得到对比结果;及
    使用得到的对比结果对所述用户身份验证结果进行修正。
  7. 根据权利要求1至6中任意一项所述的方法,其特征在于,在所述对比所述打字习惯模型和所述第一打字行为模型,得到用户身份验证结果之后,所述方法还包括:
    若所述用户身份验证结果表示用户身份是用户本人时,则
    根据所述第一打字行为模型对所述打字习惯特征模板进行再训练;及
    使用训练后的打字习惯特征模板更新未训练的打字习惯特征模板。
  8. 一种用户身份验证装置,包括:
    信息采集模块,用于采集终端录入的文本信息和所述文本信息中字符的录入间隔时间;
    模型生成模块,用于根据所述文本信息和所述录入间隔时间生成第一打字行为模型;
    模板读取模块,用于读取打字习惯特征模板;
    所述模型生成模块,还用于通过所述打字习惯特征模板将所述文本信息转换成打字习惯模型;
    模型对比模块,用于对比所述打字习惯模型和所述第一打字行为模型,得到用户身份验证结果。
  9. 根据权利要求8所述的装置,其特征在于,还包括:
    装置识别模块,用于识别用于向该终端录入该文本信息的输入装置的类型;
    指令发送模块,用于当识别到该输入装置的类型时,向该终端发送监测该类型的输入装置的监测指令;
    所述信息采集模块,还用于接收所述终端根据该监测指令采集并反馈的输入装置触发信息;
    所述模型生成模块,还用于根据所述输入装置触发信息生成第二打字行为模型;
    所述模型对比模块,还用于对比所述打字习惯模型、所述第一打字行为模型和所述第二打字行为模型,得到用户身份验证结果。
  10. 根据权利要求9所述的装置,其特征在于,还包括:
    信息解析模块,用于解析所述键位触发信息,得到被触发的键位和相应的键位触发间隔 时间;
    所述模型生成模块,还用于根据所述被触发的键位和相应的键位触发间隔时间生成所述第二打字行为模型。
  11. 一种计算机设备,包括存储器及一个或多个处理器,所述存储器中储存有计算机可读指令,所述计算机可读指令被所述一个或多个处理器执行时,使得所述一个或多个处理器执行以下步骤:
    采集终端录入的文本信息和所述文本信息中字符的录入间隔时间;
    根据所述文本信息和所述录入间隔时间生成第一打字行为模型;
    读取打字习惯特征模板;
    通过所述打字习惯特征模板将所述文本信息转换成打字习惯模型;及
    对比所述打字习惯模型和所述第一打字行为模型,得到用户身份验证结果。
  12. 根据权利要求11所述的计算机设备,其特征在于,所述处理器执行所述计算机可读指令时还执行以下步骤:
    识别用于向所述终端录入所述文本信息的输入装置的类型;
    当识别到所述输入装置的类型时,向所述终端发送监测所述类型的输入装置的监测指令;及
    接收所述终端根据所述监测指令采集并反馈的输入装置触发信息;
    所述处理器执行所述计算机可读指令时还执行以下步骤:
    根据所述输入装置触发信息生成第二打字行为模型;及
    对比所述打字习惯模型、所述第一打字行为模型和所述第二打字行为模型,得到用户身份验证结果。
  13. 根据权利要求12所述的计算机设备,其特征在于,所述处理器执行所述计算机可读指令时还执行以下步骤:
    解析所述键位触发信息,得到被触发的键位和相应的键位触发间隔时间;及
    根据所述被触发的键位和相应的键位触发间隔时间生成所述第二打字行为模型。
  14. 根据权利要求13所述的计算机设备,其特征在于,所述处理器执行所述计算机可读指令时还执行以下步骤:
    根据所述文本信息中的字符选取录入所述字符时被触发的键位;
    根据选取的键位确认相应的键位触发间隔时间;及
    对比所述键位触发间隔时间和相应的字符的录入间隔时间之间的误差是否在预设范围内;若否,则
    在生成所述第一打字行为模型和所述第二打字行为模型时,将相应的字符和相应的键位过滤。
  15. 根据权利要求12所述的计算机设备,其特征在于,所述处理器执行所述计算机可读指令时还执行以下步骤:
    监测所述虚拟键盘的布局;及
    当监测到所述布局改变时,则获取改变后的布局和改变布局花费的时间;
    所述输入装置触发信息包括虚拟键位触发信息;所述处理器执行所述计算机可读指令时还执行以下步骤:
    解析所述虚拟键位触发信息,得到被触发的虚拟键位和相应的虚拟键位触发间隔时间;
    根据所述改变布局花费的时间对所述虚拟键位触发间隔时间进行校准;及
    根据所述被触发的虚拟键位和校准后的键位触发间隔时间生成所述第二打字行为模型。
  16. 一个或多个存储有计算机可读指令的非易失性计算机可读存储介质,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器执行以下步骤:
    采集终端录入的文本信息和所述文本信息中字符的录入间隔时间;
    根据所述文本信息和所述录入间隔时间生成第一打字行为模型;
    读取打字习惯特征模板;
    通过所述打字习惯特征模板将所述文本信息转换成打字习惯模型;及
    对比所述打字习惯模型和所述第一打字行为模型,得到用户身份验证结果。
  17. 根据权利要求16所述的存储介质,其特征在于,所述处理器执行所述计算机可读指令时还执行以下步骤:
    识别用于向所述终端录入所述文本信息的输入装置的类型;
    当识别到所述输入装置的类型时,向所述终端发送监测所述类型的输入装置的监测指令;及
    接收所述终端根据所述监测指令采集并反馈的输入装置触发信息;
    所述处理器执行所述计算机可读指令时还执行以下步骤:
    根据所述输入装置触发信息生成第二打字行为模型;及
    对比所述打字习惯模型、所述第一打字行为模型和所述第二打字行为模型,得到用户身份验证结果。
  18. 根据权利要求17所述的存储介质,其特征在于,所述处理器执行所述计算机可读指令时还执行以下步骤:
    解析所述键位触发信息,得到被触发的键位和相应的键位触发间隔时间;及
    根据所述被触发的键位和相应的键位触发间隔时间生成所述第二打字行为模型。
  19. 根据权利要求18所述的存储介质,其特征在于,所述处理器执行所述计算机可读指令时还执行以下步骤:
    根据所述文本信息中的字符选取录入所述字符时被触发的键位;
    根据选取的键位确认相应的键位触发间隔时间;及
    对比所述键位触发间隔时间和相应的字符的录入间隔时间之间的误差是否在预设范围内;若否,则
    在生成所述第一打字行为模型和所述第二打字行为模型时,将相应的字符和相应的键位过滤。
  20. 根据权利要求17所述的存储介质,其特征在于,所述处理器执行所述计算机可读指 令时还执行以下步骤:
    监测所述虚拟键盘的布局;及
    当监测到所述布局改变时,则获取改变后的布局和改变布局花费的时间;
    所述输入装置触发信息包括虚拟键位触发信息;所述处理器执行所述计算机可读指令时还执行以下步骤:
    解析所述虚拟键位触发信息,得到被触发的虚拟键位和相应的虚拟键位触发间隔时间;
    根据所述改变布局花费的时间对所述虚拟键位触发间隔时间进行校准;及
    根据所述被触发的虚拟键位和校准后的键位触发间隔时间生成所述第二打字行为模型。
PCT/CN2018/089058 2018-02-08 2018-05-30 用户身份验证方法、装置、计算机设备和存储介质 WO2019153587A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810128370.1A CN108400980B (zh) 2018-02-08 2018-02-08 用户身份验证方法、装置、计算机设备和存储介质
CN201810128370.1 2018-02-08

Publications (1)

Publication Number Publication Date
WO2019153587A1 true WO2019153587A1 (zh) 2019-08-15

Family

ID=63096308

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/089058 WO2019153587A1 (zh) 2018-02-08 2018-05-30 用户身份验证方法、装置、计算机设备和存储介质

Country Status (2)

Country Link
CN (1) CN108400980B (zh)
WO (1) WO2019153587A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109635533B (zh) * 2018-12-07 2021-09-03 中国移动通信集团江苏有限公司 身份识别方法、装置、设备及介质
CN109873813B (zh) * 2019-01-28 2022-01-14 平安科技(深圳)有限公司 文本输入异常监控方法、装置、计算机设备及存储介质
US11094483B2 (en) * 2019-03-21 2021-08-17 Apple Inc. Keyboard with adjustable feedback
CN116244757A (zh) * 2023-03-15 2023-06-09 武汉天楚云计算有限公司 一种计算机设备监测警报方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150169854A1 (en) * 2013-12-16 2015-06-18 Iowa State University Research Foundation, Inc. Capturing cognitive fingerprints from keystroke dynamics for active authentication
CN105844123A (zh) * 2015-01-14 2016-08-10 中兴通讯股份有限公司 一种在终端上身份认证的方法和装置及终端
CN106203027A (zh) * 2016-06-27 2016-12-07 联想(北京)有限公司 信息处理方法及电子设备
CN106570367A (zh) * 2016-11-14 2017-04-19 北京爱知之星科技股份有限公司 基于键盘操作的用户身份识别方法和装置

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102184359B (zh) * 2011-04-29 2013-09-04 德讯科技股份有限公司 通过键盘鼠标输入习惯识别实现操作用户身份判别的方法
CN103530546B (zh) * 2013-10-25 2016-05-11 东北大学 一种基于用户鼠标行为的身份认证方法
CN104318136A (zh) * 2014-09-29 2015-01-28 同济大学 用户键盘按键行为模式建模与分析系统及其身份识别方法
US11100201B2 (en) * 2015-10-21 2021-08-24 Neurametrix, Inc. Method and system for authenticating a user through typing cadence
CN105279405B (zh) * 2015-10-28 2018-06-26 同济大学 触屏用户按键行为模式构建与分析系统及其身份识别方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150169854A1 (en) * 2013-12-16 2015-06-18 Iowa State University Research Foundation, Inc. Capturing cognitive fingerprints from keystroke dynamics for active authentication
CN105844123A (zh) * 2015-01-14 2016-08-10 中兴通讯股份有限公司 一种在终端上身份认证的方法和装置及终端
CN106203027A (zh) * 2016-06-27 2016-12-07 联想(北京)有限公司 信息处理方法及电子设备
CN106570367A (zh) * 2016-11-14 2017-04-19 北京爱知之星科技股份有限公司 基于键盘操作的用户身份识别方法和装置

Also Published As

Publication number Publication date
CN108400980A (zh) 2018-08-14
CN108400980B (zh) 2020-06-30

Similar Documents

Publication Publication Date Title
Palin et al. How do people type on mobile devices? Observations from a study with 37,000 volunteers
WO2019153587A1 (zh) 用户身份验证方法、装置、计算机设备和存储介质
US10229255B2 (en) User authentication
WO2019153604A1 (zh) 人机识别模型的建立装置、方法及计算机可读存储介质
JP6728036B2 (ja) インクからテキスト表現への変換
US20230162092A1 (en) User effort detection
Feng et al. Continuous mobile authentication using touchscreen gestures
EP2874099B1 (en) Dynamic handwriting verification and handwriting-based user authentication
CN102855082B (zh) 用于重叠文本用户输入的字符识别
US7925062B2 (en) Image processing apparatus, image processing method, signature registration program, and storage medium
CN109753653B (zh) 实体名称识别方法、装置、计算机设备和存储介质
JP2019520615A (ja) 損害賠償請求書類の文字認識方法、装置、サーバ及び記憶媒体
JP2018524734A (ja) 複数のオブジェクトの入力を認識するためのシステムならびにそのための方法および製品
CN103034429A (zh) 用于触摸屏的身份验证方法和装置
CN107273032A (zh) 信息排版方法、装置、设备及计算机存储介质
KR20210090576A (ko) 품질을 관리하는 방법, 장치, 기기, 저장매체 및 프로그램
KR20220024146A (ko) 자유 핸드라이팅 모드에서 텍스트 핸드라이팅 입력의 처리
CN103713845A (zh) 筛选候选项的方法和装置、文字输入方法和输入法系统
CN104346035A (zh) 对自动地更正的单词的指示
CN111027643A (zh) 深度神经网络模型的训练方法、人机交互识别方法、装置、电子设备及存储介质
Stylios et al. BioPrivacy: a behavioral biometrics continuous authentication system based on keystroke dynamics and touch gestures
Ellavarason et al. A framework for assessing factors influencing user interaction for touch-based biometrics
CN106250035B (zh) 动态生成个人化手写字型的系统和方法
WO2016018682A1 (en) Processing image to identify object for insertion into document
CN112163400A (zh) 信息处理方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18905412

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 03.11.2020)

122 Ep: pct application non-entry in european phase

Ref document number: 18905412

Country of ref document: EP

Kind code of ref document: A1