WO2019109433A1 - 基于步态识别的身份认证方法及装置 - Google Patents

基于步态识别的身份认证方法及装置 Download PDF

Info

Publication number
WO2019109433A1
WO2019109433A1 PCT/CN2017/120107 CN2017120107W WO2019109433A1 WO 2019109433 A1 WO2019109433 A1 WO 2019109433A1 CN 2017120107 W CN2017120107 W CN 2017120107W WO 2019109433 A1 WO2019109433 A1 WO 2019109433A1
Authority
WO
WIPO (PCT)
Prior art keywords
gait
gait cycle
acceleration data
identity authentication
preset
Prior art date
Application number
PCT/CN2017/120107
Other languages
English (en)
French (fr)
Inventor
李烨
孙方敏
毛晨飞
Original Assignee
深圳先进技术研究院
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 深圳先进技术研究院 filed Critical 深圳先进技术研究院
Publication of WO2019109433A1 publication Critical patent/WO2019109433A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/03—Arrangements for converting the position or the displacement of a member into a coded form
    • G06F3/033—Pointing devices displaced or positioned by the user, e.g. mice, trackballs, pens or joysticks; Accessories therefor
    • G06F3/0346—Pointing devices displaced or positioned by the user, e.g. mice, trackballs, pens or joysticks; Accessories therefor with detection of the device orientation or free movement in a 3D space, e.g. 3D mice, 6-DOF [six degrees of freedom] pointers using gyroscopes, accelerometers or tilt-sensors
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • G06Q50/265—Personal security, identity or safety
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/08—Network architectures or network communication protocols for network security for authentication of entities

Definitions

  • the present invention relates to the field of identity authentication technologies, and in particular, to a method and device for identity authentication based on gait recognition.
  • gait recognition based on gait recognition was first proposed by Gafurov in 2006. Through a large number of repeated experiments, he found the gait similarity between different users and the differences between different users, and concluded that all limbs are based. Sport, gait is unique and can be used for user authentication.
  • gait recognition based methods based on machine vision, based on ground sensors, based on wearable sensors. Among them, based on machine vision is to use the camera to capture a series of gait pattern images during the user's walking process, and then use image matching algorithm to achieve identity authentication, which is susceptible to light, occlusion, and distance; based on the ground sensor is passed through the carpet on the floor.
  • the force sensor captures the gait characteristics of the user and is susceptible to the external environment; the wearable sensor uses the acceleration signal to analyze the gait uniqueness.
  • the wearable sensor is the best one, but there is a problem of inaccurate detection in the prior art, because the user is walking at a fixed speed during the detection, without considering the user's The pace changes, ignoring the individual differences of users. Therefore, it is very necessary to provide a more accurate gait recognition based authentication method.
  • Another object of the present invention is to provide an identity authentication apparatus based on gait recognition to achieve more accurate identity authentication based on gait recognition.
  • an embodiment of the present invention provides a method for authenticating an identity based on gait recognition, the method comprising: acquiring an acceleration sensor disposed on a user to collect a plurality of acceleration data; and analyzing the plurality of acceleration data to extract at least a gait cycle; calculating a matching degree of the at least one gait cycle and the preset gait cycle template; and if the matching mean is greater than a preset threshold, passing the identity authentication.
  • the embodiment of the present invention further provides an identity authentication device based on gait recognition, the device comprising: an acquisition module, configured to acquire a plurality of acceleration data collected by an acceleration sensor disposed on a user; and an analysis module, And the calculating module is configured to calculate the matching degree of the at least one gait period and the preset gait period template; the comparison module is used to If the average of the matching degrees is greater than the preset threshold, the identity is authenticated.
  • the method and device for identifiable identification based on gait recognition include: acquiring a plurality of acceleration data collected by an acceleration sensor disposed on a user, and analyzing and extracting at least the plurality of acceleration sensor data
  • a gait cycle calculates an average of the matching degrees of the at least one gait cycle and the preset gait cycle template. If the matching average is greater than the preset threshold, the identity is authenticated.
  • the solution collects a plurality of acceleration data during the walking process of the user, and analyzes the plurality of acceleration data to extract a gait cycle to further match each gait cycle with the preset gait cycle template. Since different acceleration data are collected for different users and the gait cycle is calculated, the individual differences of different users are better satisfied, and the accuracy of identity recognition based on gait recognition is improved.
  • FIG. 1 is a schematic structural diagram of a terminal device according to an embodiment of the present invention.
  • FIG. 2 is a schematic flowchart diagram of a method for authenticating an identity based on gait recognition according to an embodiment of the present invention.
  • FIG. 3 is a schematic flowchart diagram of sub-steps of a gait recognition-based identity authentication method according to an embodiment of the present invention.
  • FIG. 4 is a schematic flowchart diagram of another substep of a gait recognition based identity authentication method according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram showing a unilateral spectral density function provided by an embodiment of the present invention.
  • FIG. 6 is a schematic flowchart diagram of another sub-step of a gait recognition-based identity authentication method according to an embodiment of the present invention.
  • FIG. 7 is a diagram showing an acceleration data chart provided by an embodiment of the present invention.
  • FIG. 8 is a schematic diagram showing the gait cycle detection accuracy rate provided by the embodiment of the present invention.
  • FIG. 9 is a schematic diagram showing the accuracy of identity recognition provided by an embodiment of the present invention.
  • FIG. 10 is a schematic diagram of functional modules of a gait recognition based identity authentication apparatus according to an embodiment of the present invention.
  • wearable devices In recent years, the rapid development of wearable devices has been widely used in medical, rehabilitation, interactive games and other applications. Many sensors, such as inertial sensors, magnetic sensors, and barometers, have been integrated into most wearable devices, such as smart phones or smart watches, making wearable devices smarter and more powerful. Similarly, as wearable devices become more powerful, more personal information, such as communication information, private photos, videos, etc., will be stored in the wearable device, thereby making security and privacy protection wearable. An important requirement for equipment.
  • the existing security and privacy protection for wearable devices mainly includes two methods, one is traditional password encryption and the other is biometric authentication method.
  • the traditional password encryption has the problems that the password is easy to be stolen, easy to be cracked, etc., and the wearable device is not well guaranteed; in addition, with the increase of the wearable device, the password is easily forgotten, making the user inconvenient to use.
  • Wearable device The biometric authentication method includes fingerprint recognition and face recognition, and the confidentiality is good, but the user needs to cooperate to perform encryption or decryption, and continuous authentication cannot be performed. It can be seen that the existing two methods of protecting security and privacy still have drawbacks.
  • the identity authentication based on gait recognition mainly analyzes the acceleration data collected by the acceleration sensor in the wearable device for identity authentication, and has the characteristics that the user does not need to cooperate actively and the recognition is more accurate.
  • Gait recognition mainly recognizes the identity of a person by the posture of the user walking. Since each person's gait is different, it can serve the purpose of identification. The reason is that, from a biomechanical point of view, gait is a combination of many muscles and joints of human beings, which can be described as body structure parameters. Each person's body structure is different and involves hundreds of variables, so everyone The gait is unique, so gait can be used to authenticate people and has the advantage of being more difficult to disguise and imitate to make recognition more accurate.
  • the trait-based identification authentication scheme is very widely used. It can be used for access control and security identification in special places such as prisons, airports, banks, etc. It also has potential application value in intelligent visual monitoring, and can also be used for criminal investigation of public security organs. And specific target searches, etc. If the robbers dress up, wear masks or make them easy when they rob the bank, the bank's surveillance camera can't collect the true appearance of the robbers, and even can't find the fingerprints of the robbers at the crime scene, then you can pass the robbers.
  • the walking posture is authenticated to identify the identity in order to help the police to solve the case as soon as possible.
  • FIG. 1 is a schematic structural diagram of a terminal device 100 according to an embodiment of the present invention.
  • the terminal device 100 includes an identity authentication device 110 based on gait recognition, a memory 120, a memory controller 130, a processor 140, a peripheral interface 150, an input input unit 160, an audio unit 170, a display unit 180, and a communication unit 190, wherein ,
  • the components of the memory 120, the memory controller 130, the processor 140, the peripheral interface 150, the input and output unit 160, the audio unit 170, the display unit 180, and the communication unit 190 are directly or indirectly electrically connected to each other to implement The transmission or interaction of data.
  • the components can be electrically connected to one another via one or more communication buses or signal lines.
  • the gait recognition based identity authentication device 110 includes at least one software function that can be stored in the memory 120 or in an operating system (OS) of the terminal device in the form of software or firmware. Module.
  • the processor 140 is configured to execute executable modules stored in the memory 120, such as software function modules or computer programs included in the gait recognition based identity authentication device 110.
  • the memory 120 can be, but not limited to, a random access memory (RAM), a read only memory (ROM), and a programmable read-only memory (PROM). Erasable Programmable Read-Only Memory (EPROM), Electric Erasable Programmable Read-Only Memory (EEPROM), and the like.
  • RAM random access memory
  • ROM read only memory
  • PROM programmable read-only memory
  • EPROM Erasable Programmable Read-Only Memory
  • EEPROM Electric Erasable Programmable Read-Only Memory
  • the memory 120 is configured to store a program, and the processor 140 executes the program after receiving the execution instruction, and the method performed by the terminal device defined by the flow process disclosed in any embodiment of the present invention may be applied to
  • the processor 140 is implemented by or by the processor 140.
  • Processor 140 may be an integrated circuit chip with signal processing capabilities.
  • the processor 140 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP Processor, etc.), or a digital signal processor (DSP), an application specific integrated circuit. (ASIC), Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component.
  • CPU central processing unit
  • NP Processor network processor
  • DSP digital signal processor
  • ASIC application specific integrated circuit
  • FPGA Field Programmable Gate Array
  • the general purpose processor may be a microprocessor or the processor 140 may be any conventional processor 140 or the like.
  • peripheral interface 150 couples various input/output devices to the processor 140 and to the memory 120.
  • peripheral interface 150, processor 140, and memory controller 130 can be implemented in a single chip. In other instances, they can be implemented by separate chips.
  • the input and output unit 160 is configured to provide input data to the user to implement interaction between the user and the terminal device.
  • the input and output unit 160 may be, but not limited to, a mouse, a keyboard, and the like.
  • the audio unit 170 provides an audio interface to the user, which may include one or more microphones, one or more speakers, and audio circuitry.
  • the display unit 180 provides an interactive interface (such as a user operation interface) between the terminal device and the user or for displaying image data to the user for reference.
  • the display unit 180 can be a liquid crystal display or a touch display.
  • a touch display it can be a capacitive touch screen or a resistive touch screen that supports single-point and multi-touch operations. Supporting single-point and multi-touch operations means that the touch display can sense simultaneous touch operations from one or more locations on the touch display, and the touch operation is transferred to the processor 140. Perform calculations and processing.
  • the communication unit 190 is configured to establish a connection between the wireless network and an acceleration sensor disposed on the user, so that the terminal device can send and receive data through the wireless network.
  • FIG. 2 it is a schematic flowchart of a method for authenticating an identity based on gait recognition according to an embodiment of the present invention.
  • the gait recognition-based identity authentication method provided by the solution further improves the accuracy of identity authentication, and the method includes:
  • Step S110 acquiring a plurality of acceleration data collected by an acceleration sensor disposed on the user.
  • the acceleration sensor is disposed on the user by being integrated in the wearable device.
  • the wearable device can be, but is not limited to, a mobile phone, a wristband, an anklet, etc., and the user can wear the wearable device according to his or her preference.
  • Body such as the wrist, arms, chest, waist, thighs, etc. It is easy to understand that the acceleration sensor can also be separately provided to the user.
  • the acceleration sensor disposed on the user collects a plurality of acceleration data during the walking of the user, and performs low-pass filtering on the collected acceleration data using a low-pass filter to filter out the high-frequency noise.
  • the low is low.
  • the frequency of the pass filter is set to 10 Hz.
  • the acceleration sensor can acquire three-dimensional acceleration values at the same time, that is, acceleration values in the X, Y, and Z directions. Therefore, the acceleration sensor will be further adopted.
  • the algorithm sequentially processes the acceleration values collected at the same time to obtain a plurality of acceleration data, wherein A x is an acceleration value in the X-axis direction, A z is an acceleration value in the Z-axis direction, and A y is an acceleration value in the Y-axis direction. D used as acceleration data.
  • Step S120 analyzing the plurality of acceleration data to extract at least one gait cycle.
  • step S120 includes:
  • Step S121 calculating an estimated step size according to the plurality of acceleration data.
  • the estimated step size is the length of the step by step S121 of the gait recognition based identity authentication method provided by the embodiment of the present invention.
  • Schematic diagram of the process, the step S121 includes:
  • Step S1211 performing Fourier transform on the plurality of acceleration data to obtain a one-sided spectral density function.
  • Window processing is performed on the obtained plurality of acceleration data.
  • the window size is set to 8S, and the window overlap ratio is set to 0.5.
  • the window is processed by windowing the plurality of acceleration data to perform sliding segmentation on the acceleration.
  • Fourier transform is performed on all the acceleration data. All of the acceleration data is subjected to a Fourier transform to obtain a unilateral spectral density function.
  • FIG. 5 it is a schematic diagram of a unilateral spectral density function provided by an embodiment of the present invention. Coordinates, with the spectral energy as the ordinate.
  • Step S1212 Acquire a step frequency according to the unilateral spectral density function.
  • the frequency corresponding to the value with the highest spectral energy is selected as the step frequency.
  • the selected step frequency is marked as f step .
  • Step S1213 calculating a sampling frequency of the plurality of accelerations.
  • the sampling frequency of multiple accelerations is calculated, and the sampling frequency is the number of acceleration data collected per unit time, for example, 100 acceleration data are collected within 1 s.
  • the sampling frequency is labeled f s.
  • Step S1214 calculating an estimated step size according to the step frequency and the sampling frequency.
  • the algorithm calculates the estimated step size, L e is the estimated step size, f s is the sampling frequency, and f step is the step frequency.
  • Step S122 extracting at least one gait cycle according to the plurality of acceleration data and the estimated step size.
  • the estimated step size is obtained from each user's own acceleration data analysis, the estimated step size is more reflective of each user's characteristics.
  • at least one gait cycle is extracted according to the plurality of acceleration data and the estimated step size. It is easy to understand that the more the gait cycle is extracted, the more accurate the test result is.
  • FIG. 6, which is a schematic flowchart of a sub-step of step S122 of the trait-based identification-based identity authentication method according to an embodiment of the present invention, the step S122 includes:
  • Step S1221 Perform an acceleration data chart according to the plurality of acceleration data, wherein the acceleration data chart has an amount corresponding to each acceleration as an abscissa and an acceleration as an ordinate.
  • FIG. 7 is a statistical diagram of acceleration data according to an embodiment of the present invention.
  • the acceleration data chart is drawn according to multiple acceleration data, and the number corresponding to each acceleration is the abscissa and the acceleration is the ordinate.
  • step S1222 the first minimum point on the acceleration data chart is selected as the starting point of the gait cycle.
  • the first minimum point is selected as the starting point of the gait period.
  • the starting point of the gait period is marked as sp.
  • step S1223 the estimated range of the end point of the gait cycle is determined according to the estimated step size and the starting point.
  • the estimated range of the end point of the gait cycle is determined, and the estimated range is determined by using sp+L e -d ⁇ ep ⁇ sp+L e +d.
  • L e is the estimated step size
  • sp is the starting point
  • ep is the end point of the gait cycle
  • d is the deviation of the estimated step size from the actual step size.
  • the ⁇ is preferably 0.3.
  • step S1224 the minimum value point in the estimated range is selected as the end point of the gait cycle.
  • the minimum point in the estimated range is selected as the end point ep of the gait cycle.
  • Steps S1222 to 1224 are only illustrative examples of the first gait cycle determination method. It is easy to understand that other gait cycles are also determined by the same method, such as selecting the end point ep of the current gait cycle.
  • a minimum value point is the starting point of the next gait cycle, and then the same estimation range determination method is used to determine the estimated range of the end point of the gait cycle to determine the end point of the gait cycle, and finally extract multiple gaits. cycle.
  • Step S130 normalizing each of the gait periods by using cubic spline interpolation, so that the length of the gait period is consistent with the length of the preset gait period template.
  • the lengths are normalized by cubic spline interpolation for the N gait periods.
  • each gait period after normalization is performed.
  • the length is 200, which is easy to understand.
  • the length of the gait cycle can be set according to actual needs.
  • the normalized result of H cycles is:
  • each behavior in the equation is normalized to a gait cycle.
  • the length of the gait cycle is consistent with the length of the preset gait cycle template.
  • the length of the gait cycle and the preset gait cycle template are The length is 200.
  • Step S140 calculating a matching degree of the at least one gait cycle and the preset gait cycle template.
  • the preset gait period template is established by analyzing the acceleration data of the user in advance, and the method for establishing the preset gait period template is the same as the method for extracting the gait period.
  • the preset gait cycle template is established by: acquiring a plurality of historical acceleration data collected by an acceleration sensor disposed on the user in advance, and calculating an estimated step size for the plurality of historical acceleration data, according to the plurality of historical acceleration data and the estimation
  • the step size extracts the first gait cycle, and the method of calculating the estimated step size and extracting the first gait cycle has been described above and will not be repeated here.
  • the length of the first gait cycle is normalized by cubic spline interpolation for the first gait cycle, and the first gait cycle is normalized.
  • the preset gait cycle template when the preset gait cycle template is established, only the first gait cycle is extracted after analyzing the historical acceleration data of the user, and the first gait cycle is processed to obtain the preset gait. Cycle template.
  • the reason for creating a preset gait cycle template is to extract only the first gait cycle or randomly extract a gait cycle because the template can be built faster.
  • multiple acceleration data of the user are collected, and multiple gait cycles are extracted after analyzing multiple acceleration data, and then each gait cycle is matched with the preset gait cycle template.
  • the preset gait cycle template can also analyze multiple historical acceleration data and extract multiple gait cycles, and then establish according to multiple gait cycles, such as forming an average after taking the average of multiple gait cycles. Set the gait cycle template and so on.
  • the specific implementation manner of calculating the matching degree of the at least one gait cycle and the preset gait cycle template is: calculating the Pearson correlation coefficient of each gait cycle and the preset gait cycle template respectively, specifically Algorithm to achieve, (n) is the Pearson correlation coefficient of one of the gait cycles and the preset gait cycle template. Then, calculate the mean value of the Pearson correlation coefficient of all gait cycles and the preset gait cycle template, specifically The algorithm realizes that the M score is the mean value of the Pearson correlation coefficient of all the gait cycles and the preset gait cycle template, and the mean value is the mean value of the matching degree of at least one gait cycle and the preset gait cycle template.
  • Step S150 If the average value of the matching degree is greater than a preset threshold, the identity is authenticated.
  • the preset threshold is further analyzed on the preset gait period template, and the preset threshold is fully set according to the difference of different users, and the identity authentication is more accurate through the preset threshold.
  • the matching degree is greater than the preset threshold, it is proved that the currently authenticated user has a high degree of matching with the previously entered user information, and the user identity authentication succeeds.
  • the preset gait cycle template and the preset threshold are set in advance by the scheme, and the acceleration data of the collected user is analyzed to obtain a plurality of gait cycles in actual use, and further matching is obtained according to multiple gait cycles. After the mean value, compare directly with the preset threshold to quickly reach a conclusion. It should be noted that, in this solution, whether the establishment of the preset gait cycle template or the real-time analysis of the acceleration data of the user acquires the gait cycle, it is established according to the analysis of each user's own step, and is fully considered. The individual difference of each user; in addition, the setting of the preset threshold is also set after analyzing the user's large amount of acceleration data in advance, and the preset thresholds of each user are different. Therefore, the solution greatly improves the accuracy of identity recognition based on gait recognition by fully considering the individual differences of each user and setting the gait cycle and the preset threshold.
  • the experimenter was put on a laboratory-developed sports health shirt with an acceleration sensor.
  • the acceleration sensor set on the sports health shirt was located at the chest position of the experimenter.
  • the acceleration sensor has a sampling frequency of 100 Hz and a sampling accuracy of 16 bits.
  • the sports shirts were respectively worn by three experimenters, each of which was incremented at a pace of 2 Km/h-6 Km/h and approximately 0.5 Km/h.
  • Each group of experimental personnel collected 6 sets of data each time, each set of data includes 500 gait cycles, which were collected twice, and the time interval of two data collections was 15 days.
  • the accuracy of the gait cycle extraction is detected, specifically
  • the algorithm is implemented, where N cycles is the actual number of gait cycles, and is set to 500 in the embodiment of the present invention; N detected is a gait cycle extracted by the gait recognition based identity authentication method used in actual identity authentication.
  • the number of R is the number of tests, which is 36 in the embodiment of the present invention.
  • the second aspect is to detect the accuracy of the identification, specifically Algorithm implementation, where N ii is the number of times the correct number is recognized, and N iother is the number of misidentifications.
  • N ii is the number of times the correct number is recognized
  • N iother is the number of misidentifications.
  • the identity recognition device 110 based on the gait recognition includes an obtaining module 111, an analyzing module 112, and a processing module 113. a calculation module 114 and a comparison module 115, wherein
  • the obtaining module 111 is configured to acquire a plurality of acceleration data collected by an acceleration sensor disposed on the user.
  • step S110 may be performed by the obtaining module 111.
  • the analyzing module 112 is configured to analyze the plurality of acceleration data to extract at least one gait cycle.
  • steps S120 to S1223 may be performed by the analysis module 112.
  • the processing module 113 is configured to perform normalization processing on each of the gait periods by using cubic spline interpolation, so that the length of the gait period is consistent with the length of the preset gait period template.
  • step S130 may be performed by the processing module 113.
  • the calculation module 114 is configured to calculate a matching degree of the at least one gait period and the preset gait period template.
  • step S140 may be performed by computing module 114.
  • the comparison module 115 is configured to pass identity authentication if the matching degree average is greater than a preset threshold.
  • step S150 can be performed by the comparison module 115.
  • the embodiment of the present invention further discloses a computer readable storage medium, on which a computer program is stored, and when the computer program is executed by the processor 140, the gait recognition based identity authentication method disclosed in the foregoing embodiment of the present invention is implemented.
  • the method and device for identifiable identification based on gait recognition include: acquiring a plurality of acceleration data collected by an acceleration sensor disposed on a user, and the plurality of acceleration sensors The data is analyzed to extract at least one gait cycle, and the average of the matching degree of the at least one gait cycle and the preset gait cycle template is calculated. If the average of the matching degrees is greater than the preset threshold, the identity is authenticated.
  • the solution collects a plurality of acceleration data during the walking process of the user, and analyzes the plurality of acceleration data to extract a gait cycle to further match each gait cycle with the preset gait cycle template. Since different acceleration data are collected for different users and the gait cycle is calculated, the individual differences of different users are better satisfied, and the accuracy of identity recognition based on gait recognition is improved.
  • each block of the flowchart or block diagram can represent a module, a program segment, or a portion of code that includes one or more of the Executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the blocks may also occur in a different order than those illustrated in the drawings.
  • each block of the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts can be implemented in a dedicated hardware-based system that performs the specified function or function. Or it can be implemented by a combination of dedicated hardware and computer instructions.
  • each functional module in each embodiment of the present invention may be integrated to form a separate part, or each module may exist separately, or two or more modules may be integrated to form a separate part.
  • the functions, if implemented in the form of software functional modules and sold or used as separate products, may be stored in a computer readable storage medium.
  • the technical solution of the present invention which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.
  • a computer device which may be a personal computer, server, or network device, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Human Computer Interaction (AREA)
  • Computer Security & Cryptography (AREA)
  • Signal Processing (AREA)
  • Primary Health Care (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Computing Systems (AREA)
  • Computer Hardware Design (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

本发明涉及身份认证技术领域，具体涉及一种基于步态识别的身份认证方法及装置，该方法包括：获取设置于用户身上的加速度传感器采集的多个加速度数据，对所述多个加速度传感器数据进行分析提取至少一个步态周期，计算所述至少一个步态周期与预设步态周期模板的匹配度均值，若匹配度均值大于预设阈值，则通过身份认证。本方案通过采集用户行走过程中的多个加速度数据，并对该多个加速度数据进行分析后提取步态周期，以进一步将每一个步态周期与预设步态周期模板进行匹配。由于是针对不同的用户分别采集多个加速度数据并计算步态周期，因此，更好地满足了不同用户的个体差异性，提高了基于步态识别的身份认证的准确度。

Description

基于步态识别的身份认证方法及装置
相关申请的交叉引用
本申请要求于2017年12月7日提交中国专利局的申请号为201711286270.3、名称为“基于步态识别的身份认证方法及装置”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
本发明涉及身份认证技术领域，具体而言，涉及一种基于步态识别的身份认证方法及装置。
背景技术
基于步态识别的身份认证方法最早于2006年由Gafurov提出，他通过大量重复实验发现同一用户之间的步态相似性和不同用户之间的不同之处，并得出结论：基于所有肢体的运动，步态是独一无二的，因此可用于用户的身份认证。现有技术中，基于步态识别的方法主要有三类：基于机器视觉，基于地面传感器，基于可穿戴式传感器。其中，基于机器视觉是使用摄像机捕获一系列用户走路过程中的步态模式图像，然后使用图像匹配算法实现身份认证，易受光、遮挡、距离的影响；基于地面传感器是通过地板上地毯内中的力传感器捕获用户的步态特征，易受外部环境的影响；基于可穿戴式传感器采用加速度信号来分析步态唯一性。在三种现有技术中，基于可穿戴式传感器是最好的一种，但是在现有技术中存在检测不精准的问题，其原因在于检测时使用户以固定速度行走，没有考虑到用户的步速变化，忽略了用户的个体差异性。因此，提供一种更精确的基于步态识别的身份认证方法是十分有必要的。
发明内容
本发明的目的在于提供一种基于步态识别的身份认证方法，以实现更精准地基于步态识别进行身份认证。
本发明的另一目的在于提供一种基于步态识别的身份认证装置，以实现更精准地基于步态识别进行身份认证。
为了实现上述目的，本发明实施例采用的技术方案如下：
第一方面，本发明实施例提供了种基于步态识别的身份认证方法，所述方法包括：获取设置于用户身上的加速度传感器采集多个加速度数据；对所述多个加速度数据进行分析提取至少一个步态周期；计算所述至少一个步态周期与预设步态周期模板的匹配度均值；若所述匹配度均值大于预设阈值，则通过身份认证。
第二方面，本发明实施例还提供了一种基于步态识别的身份认证装置，所述装置包括：获取模块，用于获取设置于用户身上的加速度传感器采集的多个加速度数据；分析模块，用于对所述多个加速度数据进行分析提取至少一个步态周期；计算模块，用于计算所述至少一个步态周期与预设步态周期模板的匹配度均值；比较模块，用于若所述匹配度均值大于预设阈值，则通过身份认证。
本发明实施例提供的一种基于步态识别的身份认证方法及装置，该方法包括：获取设置于用户身上的加速度传感器采集的多个加速度数据，对所述多个加速度传感器数据进行分析提取至少一个步态周期，计算所述至少一个步态周期与预设步态周期模板的匹配度均值，若匹配度均值大于预设阈值，则通过身份认证。本方案通过采集用户行走过程中的多个加速度数据，并对该多个加速度数据进行分析后提取步态周期，以进一步将每一个步态周期与预设步态周期模板进行匹配。由于是针对不同的用户分别采集多个加速度数据并计算步态周期，因此，更好地满足了不同用户的个体差异性，提高了基于步态识别的身份认证的准确度。
为使本发明的上述目的、特征和优点能更明显易懂，下文特举较佳实施例，并配合所附附图，作详细说明如下。
附图说明
为了更清楚地说明本发明实施例的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，应当理解，以下附图仅示出了本发明的某些 实施例，因此不应被看作是对范围的限定，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他相关的附图。
图1示出了本发明实施例提供的一种终端设备的结构示意图。
图2示出了本发明实施例提供的一种基于步态识别的身份认证方法的流程示意图。
图3示出了本发明实施例提供的一种基于步态识别的身份认证方法的子步骤的流程示意图。
图4示出了本发明实施例提供的一种基于步态识别的身份认证方法的另一子步骤的流程示意图。
图5示出了本发明实施例提供的一种单边频谱密度函数的示意图。
图6示出了本发明实施例提供的一种基于步态识别的身份认证方法的另一子步骤的流程示意图。
图7示出了本发明实施例提供的一种加速度数据统计图。
图8示出了本发明实施例提供的步态周期检测准确率的示意图。
图9示出了本发明实施例提供的身份识别准确率的示意图。
图10示出了本发明实施例提供的种基于步态识别的身份认证装置的功能模块示意图。
图示：100-终端设备；110-基于步态识别的身份认证装置；120-存储器；130-存储控制器；140-处理器；150-外设接口；160-输入输出单元；170-音频单元；180-显示单元；190-通信单元；111-获取模块；112-分析模块；113-处理模块；114-计算模块；115-比较模块。
具体实施方式
下面将结合本发明实施例中附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。通常在此处附图中描述和示出的本发明实施例的组件可以以各种不同的配置来布置和设计。因此，以下对在附图中提供的本发明的实施例的详细描述并非旨在限制要求保护的本发明的范围，而是仅仅表示本发明的选定实施例。基于本发明的实施例，本领域技术人员在没 有做出创造性劳动的前提下所获得的所有其他实施例，都属于本发明保护的范围。
应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步定义和解释。同时，在本发明的描述中，术语“第一”、“第二”等仅用于区分描述，而不能理解为指示或暗示相对重要性。
近年来穿戴式设备的快速发展，已广泛地应用于医疗、康复、互动游戏等多种应用领域。许多传感器，如惯性传感器、磁传感器、气压高度计已经集成在大多数穿戴式设备(如智能电话或智能手表等)中，使得穿戴式设备的更加智能和强大。同样地，随着穿戴式设备的功能越来越强大，则该穿戴式设备内将存储更多的私有信息，如通信信息、私人照片、视频等，由此，安全和隐私保护成为了穿戴式设备的一个重要需求。现有的针对穿戴式设备的安全和隐私保护主要包括两种方法，一种为传统的密码加密，另一种为生物认证方法。其中，传统的密码加密存在密码易被盗取、易被破解等问题，不能很好地保证穿戴式设备的安全；此外，随着穿戴式设备的增多，密码容易被忘记，使得用户不便于使用穿戴式设备。该生物认证方法包括指纹识别和人脸识别，其保密性较好，但需要用户配合才能进行加密或解密，且不能进行连续认证。由此可见，现有的两种对安全和隐私进行保护的方法仍然存在缺陷。
对于基于步态识别的身份认证主要是对穿戴式设备内的加速度传感器采集的加速度数据进行分析以进行身份认证，其具有不需要用户主动配合且识别更加精准的特性。步态识别主要是通过用户走路的姿势来识别人的身份，由于每个人的步态是不一样的，故可以起到身份识别的目的。其原因在于，从生物力学的角度解释，步态是人的众多肌肉和关节的组合运动，可描述成身体结构参数，每个人的身体结构各不相同且涉及数百个变量，因此，每个人的步态都是唯一的，因此，步态可以用于对人进行身份认证，且具有因难以伪装和模仿以使得识别更准确的优势。
基于步态识别的身份认证使用场景非常广，其可用于监狱、机场、银 行等特殊场所的访问控制和安全鉴定，在智能视觉监控方面也具有潜在的应用价值，还可以用于公安机关刑事侦查以及特定目标搜索等。如若劫匪在抢劫银行时进行化妆、戴面具或易容等装扮，则银行的监控摄像头无法采集劫匪的真实容貌，甚至在犯罪现场也不能找到劫匪的指纹，则此时可通过劫匪的行走姿势进行身份认证，以辨别身份，以便帮助警察尽快破案。
本发明实施例提供的一种基于步态识别的身份认证方法应用于终端设备，该终端设备可以是，但不限于，电脑等智能电子设备。请参照图1，是本发明实施例提供的一种终端设备100的结构示意图。该终端设备100包括基于步态识别的身份认证装置110、存储器120、存储控制器130、处理器140、外设接口150、输入输入单元160、音频单元170、显示单元180以及通信单元190，其中，
所述存储器120、存储控制器130、处理器140、外设接口150、输入输出单元160、音频单元170、显示单元180、通信单元190各元件相互之间直接或间接地电性连接，以实现数据的传输或交互。例如，这些元件相互之间可通过一条或多条通讯总线或信号线实现电性连接。所述基于步态识别的身份认证装置110包括至少一个可以软件或固件(firmware)的形式存储于所述存储器120中或固化在所述终端设备的操作系统(operating system，OS)中的软件功能模块。所述处理器140用于执行存储器120中存储的可执行模块，例如基于步态识别的身份认证装置110包括的软件功能模块或计算机程序。
其中，存储器120可以是，但不限于，随机存取存储器(Random Access Memory，RAM)，只读存储器(Read Only Memory，ROM)，可编程只读存储器(Programmable Read-Only Memory，PROM)，可擦除只读存储器(Erasable Programmable Read-Only Memory，EPROM)，电可擦除只读存储器(Electric Erasable Programmable Read-Only Memory，EEPROM)等。其中，存储器120用于存储程序，所述处理器140在接收到执行指令后，执行所述程序， 后续本发明实施例任一实施例揭示的流过程定义的终端设备所执行的方法可以应用于处理器140中，或者由处理器140实现。
处理器140可能是一种集成电路芯片，具有信号的处理能力。上述的处理器140可以是通用处理器，包括中央处理器(Central Processing Unit，简称CPU)、网络处理器(Network Processor，简称NP)等；还可以是数字信号处理器(DSP)、专用集成电路(ASIC)、现场可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本发明实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器140也可以是任何常规的处理器140等。
所述外设接口150将各种输入/输出装置耦合至处理器140以及存储器120。在一些实施例中，外设接口150，处理器140以及存储控制器130可以在单个芯片中实现。在其他一些实例中，他们可以分别由独立的芯片实现。
输入输出单元160用于提供给用户输入数据实现用户与终端设备的交互。所述输入输出单元160可以是，但不限于，鼠标和键盘等。
音频单元170向用户提供音频接口，其可包括一个或多个麦克风、一个或者多个扬声器以及音频电路。
显示单元180在终端设备与用户之间提供一个交互界面(例如用户操作界面)或用于显示图像数据给用户参考。在本实施例中，所述显示单元180可以是液晶显示器或触控显示器。若为触控显示器，其可为支持单点和多点触控操作的电容式触控屏或电阻式触控屏等。支持单点和多点触控操作是指触控显示器能感应到来自该触控显示器上一个或多个位置处同时产生的触控操作，并将该感应到的触控操作交由处理器140进行计算和处理。
所述通信单元190用于通过所述无线网络与设置于用户身上的加速度传感器之间建立连接，从而实现所述终端设备通过无线网络收发数据。
请参照图2，是本发明实施例提供的一种基于步态识别的身份认证方法的流程示意图。通过本方案提供的基于步态识别的身份认证方法以进一步提高了身份认证的准确度，该方法包括：
步骤S110，获取设置于用户身上的加速度传感器采集的多个加速度数据。
该加速度传感器通过集成于穿戴式设备中设置于用户身上，如该穿戴式设备可以是，但不限于，手机、手环、脚链等，则用户可以根据自己的喜好将该穿戴式设备佩戴于身上，如腕部、手臂、胸部、腰部、大腿等。容易理解的，该加速度传感器也可单独设置于用户身上。进而，通过该设置于用户身上的加速度传感器采集用户行走过程中的多个加速度数据，且对采集的加速度数据使用低通滤波器进行低通滤波以滤除高频噪声，较优地，该低通滤波器的频率设置为10HZ。需要说明的是，该加速度传感器可在同一时刻采集三维的加速度值，即X、Y、Z方向的加速度值，因此，该加速度传感器将进一步采用
Figure PCTCN2017120107-appb-000001
算法依次对同一时刻采集的加速度值进行处理，以得到多个加速度数据，其中，A x为X轴方向的加速度值，A z为Z轴方向的加速度值，A y为Y轴方向的加速度值，D used为加速度数据。
步骤S120，对所述多个加速度数据进行分析提取至少一个步态周期。
由于该多个加速度数据是针对不同的用户分别采集的，则对每个用户的多个加速度数据分别进行分析提取至少一个步态周期，由于每个用户走路的姿态、速度、步长等均不同，则该步态周期是唯一的，不同的用户对应不同的步态周期。请参照图3，是本发明实施例提供的一种基于步态识别的身份认证方法的步骤S120的子步骤的流程示意图，该步骤S120包括：
步骤S121，根据所述多个加速度数据计算预估步长。
该预估步长为用户每一步跨出的长度，该预估步长的计算方法请参照图4是本发明实施例提供的一种基于步态识别的身份认证方法的步骤 S121的子步骤的流程示意图，该步骤S121包括：
步骤S1211，对所述多个加速度数据进行傅里叶变换得到单边频谱密度函数。
对获取的多个加速度数据进行加窗处理，较优地，将该窗口大小设置为8S，窗口重叠率设置为0.5，通过对多个加速度数据进行加窗处理，以对加速度进行滑动分段，以便于对每段加窗的加速度数据分别进行傅里叶变换，进而对所有的加速度数据进行傅里叶变换。所有加速度数据经过傅里叶变换后得到单边频谱密度函数，如图5所示，是本发明实施例提供的一种单边频谱密度函数的示意图，该单边频谱密度函数示意图以频率为横坐标，以频谱能量为纵坐标。
步骤S1212，根据所述单边频谱密度函数获取步频。
选取频谱能量最高的值对应的频率为步频，在本发明实施例中，将选取的步频标记为f step。
步骤S1213，计算所述多个加速度的采样频率。
由于需要对同一个用户采集多个加速度数据，则计算多个加速度的采样频率，该采样频率为单位时间内采集的加速度数据的个数，如1s内采集了100个加速度数据。在本发明实施例中，将采样频率标记为f s。
步骤S1214，根据所述步频和采样频率计算预估步长。
根据步频和采样频率计算预估步长，具体为，采用
Figure PCTCN2017120107-appb-000002
算法计算预估步长，L e为预估步长，f s为采样频率，f step为步频。
步骤S122，根据所述多个加速度数据和所述预估步长提取至少一个步态周期。
由于预估步长是根据每个用户自身的加速度数据分析获得，因此，该预估步长更能反映每个用户的特性。鉴于此，再根据所述多个加速度数据和预估步长提取至少一个步态周期，容易理解的，步态周期提取得越多，检验结果就更加精确。请参照图6，是本发明实施例提供的一种基于步态识别的身份认证方法的步骤S122的子步骤的流程示意图，该步骤S122包括：
步骤S1221，根据所述多个加速度数据作出加速度数据统计图，所述加速度数据统计图以每一个加速度对应的数量为横坐标，以加速度为纵坐标。
请参照图7，是本发明实施例提供的一种加速度数据统计图，该加速度数据统计图根据多个加速度数据进行绘制，其以每一个加速度对应的数量为横坐标，以加速度为纵坐标。
步骤S1222，选取所述加速度数据统计图上第一个极小值点为步态周期的起始点。
在图7所示的加速度数据统计图中，选取第一个极小值点为步态周期的起始点，在本发明实施例中，该步态周期的起始点标记为sp。
步骤S1223，根据预估步长和起始点确定步态周期的终点所在的预估范围。
根据预估步长L e和起始点sp确定步态周期的终点所在的预估范围，具体采用sp+L e-d＜ep＜sp+L e+d方式进行确定预估范围。其中，L e为预估步长，sp为起始点，ep为步态周期的终点，d为预估步长与实际步长的偏差值。需要说明的是，d可以通过d＝βL e＝0.3×L e算法进行确定，其中L e为预估步长，β为预估步长的误差，在本发明实施例中，该β优选为0.3。通过对步态周期的终点ep限定到该预估范围内，使得步态周期的终点的正确性和唯一性。
步骤S1224，选取所述预估范围内的极小值点为步态周期的终点。
在图7所示的加速度数据统计图中选取预估范围内的极小值点为步态周期的终点ep。步骤S1222～步骤1224只是示例性地举出了第一个步态周期的确定方法，容易理解的，其他步态周期也采用同样的方法进行确定，如选取当前步态周期的终点ep后的第一个极小值点为下一个步态周期的起始点，然后采用同样的预估范围确定方法确定步态周期的终点的预估范围进而确定步态周期的终点，最终提取得到多个步态周期。
步骤S130，采用三次样条插值对每个所述步态周期进行归一化处理，以使所述步态周期的长度与预设步态周期模板的长度一致。
若提取得到N个步态周期，则对该N个步态周期分别采用三次样条插值对长度进行归一化处理，在发明实施例中，每个进行归一化处理后的步态周期的长度为200，容易理解的，该步态周期的长度可根据实际需要进行设置。该归一化的结果H cycles为：
Figure PCTCN2017120107-appb-000003
该算式中的每一行为归一化后的一个步态周期。通过对每一个步态周期进行归一化处理使得步态周期的长度与预设步态周期模板的长度一致，在本发明实施例中，该步态周期的长度与预设步态周期模板的长度均为200。
步骤S140，计算所述至少一个步态周期与预设步态周期模板的匹配度均值。
在本发明实施例中，该预设步态周期模板为事先采集用户的加速度数据进行分析后建立，该预设步态周期模板的建立方法与步态周期的提取方法相同。该预设步态周期模板的建立过程为：事先获取设置于用户身上的加速度传感器采集的多个历史加速度数据，对多个历史加速度数据计算预估步长，根据多个历史加速度数据和预估步长提取第一个步态周期，其计算预估步长和提取第一个步态周期的方法已经在前面说明，在此不再重复。
提取到第一个步态周期后，对该第一个步态周期采用三次样条插值对该第一个步态周期的长度进行归一化处理，该第一个步态周期经归一化处理后的长度为200，进而将经归一化处理后的第一个步态周期保存为能够表征该用户身份的预设步态周期模板，即T temp＝spline(T cycle)，其中，T temp为预设步态周期模板，T cycle为经归一化处理后的第一个步态周期。
需要说明的是，建立预设步态周期模板时，对用户的多个历史加速度数据进行分析后只提取了第一个步态周期，对第一个步态周期进行处理后得到预设步态周期模板。之所以在建立预设步态周期模板时，只提取第一个步态周期或随意提取一个步态周期，是因为能够更快地建立模板。同时， 由于在实际应用时，采集了用户多个加速度数据，且对多个加速度数据分析后提取了多个步态周期，然后分别将每一个步态周期均与预设步态周期模板进行匹配，使得实际应用时，具有较高的准确性。容易理解的，该预设步态周期模板也可对多个历史加速度数据进行分析后提取多个步态周期，进而依据多个步态周期建立，如对多个步态周期取均值后形成预设步态周期模板等。
其计算所述至少一个步态周期与预设步态周期模板的匹配度均值的具体实现方式为：将每一个步态周期分别与预设步态周期模板进行皮尔逊相关系数计算，具体通过
Figure PCTCN2017120107-appb-000004
算法来实现，
Figure PCTCN2017120107-appb-000005
(n)为其中一个步态周期与预设步态周期模板的皮尔逊相关系数。然后，计算所有步态周期与预设步态周期模板的皮尔逊相关系数的均值，具体通过
Figure PCTCN2017120107-appb-000006
算法实现，M score为所有步态周期与预设步态周期模板的皮尔逊相关系数的均值，该均值即为至少一个步态周期与预设步态周期模板的匹配度均值。
步骤S150，若所述匹配度均值大于预设阈值，则通过身份认证。
该预设阈值在预设步态周期模板上进一步分析设置，该预设阈值充分考虑不同用户的差异性进行设置，进而通过预设阈值进行身份认证更加准确。该预设阈值的设置方式为：将事先采集的历史加速度数据Dused＝{xi,i＝1,2,…L}分为两部分，一部分为用于分析第一个步态周期，即预设步态周期模板的历史加速度数据，表征为Tcycle＝{xi,i＝sp,sp+1……ep}，另一部分为所有历史加速度数据除去用于预设步态周期模板部分的其他加速度数据，表征为Drest＝{xj,j＝ep+1,ep+2,……L}。将Drest分割成与Tcycle等长的数据集，即假设Tcycle的长度为s＝ep-sp+1,将Drest分割为k个长度均为s的数据集Ds(k)，该Ds(k)表示为D s(k)＝[D rest(k)，…D rest(k+s)]，k＝1，2，3……R。进而计算Tcycle和每个Ds(k)的皮尔逊相关系数ρ(k)，该ρ(k)的算法为：ρ(k)＝PCC(T cycle，D s(k))，k＝1.2....R。最后通过算法
Figure PCTCN2017120107-appb-000007
计算出预设阈值Th。
若该匹配度均值大于预设阈值，则证明当前验证的用户与事先录入的用户信息匹配程度高，则该用户身份认证成功。
由此可见，通过本方案事先设置预设步态周期模板以及预设阈值，在实际使用时，对采集的用户的加速度数据进行分析获取多个步态周期，进一步根据多个步态周期获取匹配度均值后，直接与预设阈值进行比较，以快速得出结论。需要说明的是，由于本方案中无论是预设步态周期模板的建立还是实时对用户的加速度数据分析获取步态周期，均是根据每个用户的自身的步长分析后建立，充分考虑了每个用户的个体差异；此外，预设阈值的设置也是事先对用户的大量加速度数据进行分析后进行设置，每个用户的预设阈值均不相同。因此，本方案通过充分考虑每个用户的个体差异性后设置步态周期和预设阈值，以极大地提高了基于步态识别的身份认证的准确率。
为了更好地说明本发明实施例提供的一种基于步态识别的身份认证方法具有较一般的步态识别方法更高的准确率，以下通过一个实验说明。具体如下：
让实验人员穿上实验室开发的带有加速度传感器的运动健康衫，该运动健康衫上设置的加速度传感器位于实验人员的胸部位置，该加速度传感器的采样频率为100HZ，采样精度为16位。
将该运动健康衫分别让三个实验人员穿，每个实验人员以步速2Km/h-6Km/h且大致以0.5Km/h的幅度递增。对每个实验人员每人每次采集6组数据，每组数据包括500个步态周期，共采集两次，两次数据采集的时间间隔为15天。
对采集的数据进行两个方面的分析，具体为：
第一方面，对步态周期提取的准确率进行检测，具体通过
Figure PCTCN2017120107-appb-000008
算法实现，其中，N cycles为实际的步态周期数，在本发明实施例中，设置为500；N detected为在实际进行身份认证时采用的基于步态识别的身份认证方法提取的步态周期数；R为测试次数，在本发明实施例中，为36。通过对实际的步态周期数和方法算出的步态周期数的分 析得到如图8所示的步态周期检测准确率的示意图，由图可知，通过算法提取出的步态周期准确率极高。
第二方面，对身份识别准确率进行检测，具体通过
Figure PCTCN2017120107-appb-000009
算法实现，其中，N ii为识别正确的次数，N iother为误识别的次数。在本发明实施例中，通过
Figure PCTCN2017120107-appb-000010
计算出3个实验人员的身份识别准确率，如图9所示的身份识别准确率的示意图，由图可知，通过本方法是身份识别准确率极高，约高于现有的步态识别方法的5.7％。因此，通过本方案提供的一种基于步态识别的身份认证方法能极大地提高身份认证的准确率。
请参照图10，是本发明实施例提供的一种基于步态识别的身份认证装置110的功能模块示意图，该基于步态识别的身份认证装置110包括获取模块111、分析模块112、处理模块113、计算模块114以及比较模块115，其中，
获取模块111，用于获取设置于用户身上的加速度传感器采集的多个加速度数据。
在本发明实施例中，步骤S110可以由获取模块111执行。
分析模块112，用于对所述多个加速度数据进行分析提取至少一个步态周期。
在本发明实施例中，步骤S120～S1223可以由分析模块112执行。
处理模块113，用于采用三次样条插值对每个所述步态周期进行归一化处理，以使所述步态周期的长度与预设步态周期模板的长度一致。
在本发明实施例中，步骤S130可以由处理模块113执行。
计算模块114，用于计算所述至少一个步态周期与预设步态周期模板的匹配度均值。
在本发明实施例中，步骤S140可以由计算模块114执行。
比较模块115，用于若所述匹配度均值大于预设阈值，则通过身份认证。
在本发明实施例中，步骤S150可以由比较模块115执行。
由于在基于步态识别的身份认证方法部分已经详细描述，在此不再赘 述。
本发明实施例还揭示了一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器140执行时实现前述本发明实施例揭示的基于步态识别的身份认证方法。
综上所述，本发明实施例提供的一种基于步态识别的身份认证方法及装置，该方法包括：获取设置于用户身上的加速度传感器采集的多个加速度数据，对所述多个加速度传感器数据进行分析提取至少一个步态周期，计算所述至少一个步态周期与预设步态周期模板的匹配度均值，若匹配度均值大于预设阈值，则通过身份认证。本方案通过采集用户行走过程中的多个加速度数据，并对该多个加速度数据进行分析后提取步态周期，以进一步将每一个步态周期与预设步态周期模板进行匹配。由于是针对不同的用户分别采集多个加速度数据并计算步态周期，因此，更好地满足了不同用户的个体差异性，提高了基于步态识别的身份认证的准确度。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置和方法，也可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，附图中的流程图和框图显示了根据本发明的多个实施例的装置、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或代码的一部分，所述模块、程序段或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现方式中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
另外，在本发明各个实施例中的各功能模块可以集成在一起形成一个独立的部分，也可以是各个模块单独存在，也可以两个或两个以上模块集成形成一个独立的部分。
所述功能如果以软件功能模块的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
以上所述仅为本发明的优选实施例而已，并不用于限制本发明，对于本领域的技术人员来说，本发明可以有各种更改和变化。凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步定义和解释。
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可 轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应所述以权利要求的保护范围为准。

Claims (10)

  1. 一种基于步态识别的身份认证方法，其特征在于，所述方法包括如下步骤：
    获取设置于用户身上的加速度传感器采集的多个加速度数据；
    对所述多个加速度数据进行分析提取至少一个步态周期；
    计算所述至少一个步态周期与预设步态周期模板的匹配度均值；
    若所述匹配度均值大于预设阈值，则通过身份认证。
  2. 如权利要求1所述的基于步态识别的身份认证方法，其特征在于，所述对所述多个加速度数据进行分析提取至少一个步态周期的步骤包括：
    根据所述多个加速度数据计算预估步长；
    根据所述多个加速度数据和所述预估步长提取至少一个步态周期。
  3. 如权利要求2所述的基于步态识别的身份认证方法，其特征在于，所述根据所述多个加速度数据计算预估步长的步骤包括：
    对所述多个加速度数据进行傅里叶变换得到单边频谱密度函数；
    根据所述单边频谱密度函数获取步频；
    计算所述多个加速度的采样频率；
    根据所述步频和采样频率计算预估步长。
  4. 如权利要求2所述的基于步态识别的身份认证方法，其特征在于，所述根据所述多个加速度数据和所述预估步长提取至少一个步态周期的步骤包括：
    根据所述多个加速度数据作出加速度数据统计图，所述加速度统计图以每一个加速度对应的数量为横坐标，以加速度为纵坐标；
    选取所述加速度数据统计图上第一个极小值点为步态周期的起始点；
    根据预估步长和起始点确定步态周期的终点所在的预估范围；
    选取所述预估范围内的极小值点为步态周期的终点。
  5. 如权利要求1所述的基于步态识别的身份认证方法，其特征在于，所述计算所述至少一个步态周期与预设步态周期模板的匹配度均值的步 骤包括：
    计算每个所述步态周期与预设步态周期模板的皮尔逊相关系数；
    计算所有所述皮尔逊相关系数的均值，所述皮尔逊相关系数的均值为所述至少一个步态周期与预设步态周期模板的匹配度均值。
  6. 如权利要求1所述的基于步态识别的身份认证方法，其特征在于，所述对所述多个加速度数据进行分析提取至少一个步态周期之后还包括步骤：
    采用三次样条插值对每个所述步态周期进行归一化处理，以使所述步态周期的长度与预设步态周期模板的长度一致。
  7. 如权利要求1所述的基于步态识别的身份认证方法，其特征在于，所述预设步态周期模板为获取事先设置于用户的加速度传感器采集的多个历史加速度数据，以依据对所述多个历史加速度数据分析后确定的第一个步态周期而建立，对所述多个历史加速度数据分析包括对所述多个历史加速度数据进行傅里叶变换得到单边频谱密度函数，根据所述单边频谱密度函数获取的步频和采样频率计算预估步长，根据预估步长确定第一个步态周期，并对所述第一个步态周期进行归一化处理。
  8. 如权利要求7所述的基于步态识别的身份认证方法，其特征在于，所述预设阈值通过将所述多个历史加速度数据中除第一个步态周期以外的其他历史加速度数据分割成多个预定长度的数据集，计算每个数据集与预设步态周期模板的皮尔逊相关系数，并根据多个皮尔逊相关系数计算得到。
  9. 一种基于步态识别的身份认证装置，其特征在于，所述装置包括：
    获取模块，配置成获取设置于用户身上的加速度传感器采集的多个加速度数据；
    分析模块，配置成对所述多个加速度数据进行分析提取至少一个步态周期；
    计算模块，配置成计算所述至少一个步态周期与预设步态周期模板的匹配度均值；
    比较模块，配置成若所述匹配度均值大于预设阈值，则通过身份认证。
  10. 如权利要求9所述的基于步态识别的身份认证装置，其特征在于，所述装置还包括：
    处理模块，配置成采用三次样条插值对每个所述步态周期进行归一化处理，以使所述步态周期的长度与预设步态周期模板的长度一致。
PCT/CN2017/120107 2017-12-07 2017-12-29 基于步态识别的身份认证方法及装置 WO2019109433A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711286270.3 2017-12-07
CN201711286270.3A CN107831907A (zh) 2017-12-07 2017-12-07 基于步态识别的身份认证方法及装置

Publications (1)

Publication Number Publication Date
WO2019109433A1 true WO2019109433A1 (zh) 2019-06-13

Family

ID=61641888

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/120107 WO2019109433A1 (zh) 2017-12-07 2017-12-29 基于步态识别的身份认证方法及装置

Country Status (2)

Country Link
CN (1) CN107831907A (zh)
WO (1) WO2019109433A1 (zh)

Families Citing this family (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108596100A (zh) * 2018-04-25 2018-09-28 常州大学 一种基于多姿态下的人员识别方法
CN108563939B (zh) * 2018-04-25 2022-05-20 常州大学 基于步态轨迹曲线特征的人体身份识别
CN108932504A (zh) * 2018-07-24 2018-12-04 中国科学院深圳先进技术研究院 身份认证方法、装置、电子设备及存储介质
WO2020019171A1 (zh) * 2018-07-24 2020-01-30 中国科学院深圳先进技术研究院 身份认证方法、装置、电子设备及存储介质
CN109124646B (zh) * 2018-09-26 2021-06-18 北京壹氢科技有限公司 一种适用于配带智能手机行人的步态检测方法
CN109302532B (zh) * 2018-11-02 2020-02-07 北京航空航天大学 一种基于智能手机加速度传感器的身份认证方法
EP3711666A1 (en) * 2019-03-21 2020-09-23 Koninklijke Philips N.V. Analysing movement of a subject
CN110017834B (zh) * 2019-04-15 2021-12-24 歌尔科技有限公司 使用对象确定方法、设备及存储介质
CN111126294B (zh) * 2019-12-25 2023-06-16 每日互动股份有限公司 一种基于移动终端数据识别终端用户步态的方法和服务器
CN111142688B (zh) * 2019-12-25 2023-05-12 每日互动股份有限公司 一种基于移动终端数据识别终端用户换机的方法和服务器
CN111062352B (zh) * 2019-12-25 2023-07-14 每日互动股份有限公司 一种基于移动终端数据识别终端用户步态的方法和服务器
CN111061376B (zh) * 2019-12-25 2023-07-18 每日互动股份有限公司 一种基于移动终端数据识别终端用户换机的方法和服务器
CN110990819B (zh) * 2019-12-25 2023-04-21 每日互动股份有限公司 一种基于移动终端数据获取终端用户步态特征数据的方法和服务器
CN111062353B (zh) * 2019-12-25 2023-04-28 每日互动股份有限公司 一种基于移动终端数据获取终端用户步态特征数据的方法和服务器
CN111214236A (zh) * 2020-01-14 2020-06-02 珠海格力电器股份有限公司 监护方法、装置、设备及存储介质
CN112257680B (zh) * 2020-11-23 2023-06-23 中国联合网络通信集团有限公司 脑纹身份认证识别方法、服务器及计算机可读存储介质
CN112580759A (zh) * 2020-12-16 2021-03-30 珠海格力电器股份有限公司 一种储物方法、储物装置及储物柜
CN113296608A (zh) * 2021-05-28 2021-08-24 深圳中科健安科技有限公司 一种渐进式智能督查方法及系统
CN114287921A (zh) * 2021-12-23 2022-04-08 常州信息职业技术学院 一种步态双侧相似性分析方法、装置及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101695445A (zh) * 2009-10-29 2010-04-21 浙江大学 一种基于加速度传感器的步态身份识别方法
US9188963B2 (en) * 2009-07-06 2015-11-17 Autonomous Id Canada Inc. Gait-based authentication system
CN106056013A (zh) * 2016-06-03 2016-10-26 深圳市金立通信设备有限公司 一种终端防盗方法及终端
CN107103297A (zh) * 2017-04-20 2017-08-29 武汉理工大学 基于手机加速度传感器的步态身份识别方法及系统
CN107122704A (zh) * 2017-03-16 2017-09-01 华南理工大学 一种基于运动传感器的步态识别方法
CN107273726A (zh) * 2017-06-02 2017-10-20 中国人民解放军信息工程大学 基于加速度周期变化规律的设备机主身份实时识别方法及其装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103796233B (zh) * 2014-03-07 2017-03-29 武汉虹信通信技术有限责任公司 无线网络空口参数的室内自动地理化展示方法及系统
CN103983273B (zh) * 2014-04-29 2017-06-06 华南理工大学 一种基于加速度传感器的实时步长估计方法
CN106595912A (zh) * 2016-11-01 2017-04-26 中国海洋大学 一种人足动态力学实时检测评估系统及方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9188963B2 (en) * 2009-07-06 2015-11-17 Autonomous Id Canada Inc. Gait-based authentication system
CN101695445A (zh) * 2009-10-29 2010-04-21 浙江大学 一种基于加速度传感器的步态身份识别方法
CN106056013A (zh) * 2016-06-03 2016-10-26 深圳市金立通信设备有限公司 一种终端防盗方法及终端
CN107122704A (zh) * 2017-03-16 2017-09-01 华南理工大学 一种基于运动传感器的步态识别方法
CN107103297A (zh) * 2017-04-20 2017-08-29 武汉理工大学 基于手机加速度传感器的步态身份识别方法及系统
CN107273726A (zh) * 2017-06-02 2017-10-20 中国人民解放军信息工程大学 基于加速度周期变化规律的设备机主身份实时识别方法及其装置

Also Published As

Publication number Publication date
CN107831907A (zh) 2018-03-23

Similar Documents

Publication Publication Date Title
WO2019109433A1 (zh) 基于步态识别的身份认证方法及装置
Kang et al. ECG authentication system design based on signal analysis in mobile and wearable devices
Juefei-Xu et al. Gait-id on the move: Pace independent human identification using cell phone accelerometer dynamics
CN105787420B (zh) 用于生物认证的方法、装置以及生物认证系统
Rong et al. Identification of individual walking patterns using gait acceleration
Trung et al. Performance evaluation of gait recognition using the largest inertial sensor-based gait database
US20180068193A1 (en) Method, Apparatus And System For Biometric Identification
CN107437074B (zh) 一种身份认证方法和装置
US20160183812A1 (en) Biometric authentication method and apparatus
Sarkar et al. Deep feature-based face detection on mobile devices
Chen et al. Your face your heart: Secure mobile face authentication with photoplethysmograms
Ayeswarya et al. A survey on different continuous authentication systems
Sharma et al. A new pose invariant face recognition system using PCA and ANFIS
WO2018152711A1 (zh) 一种基于心电认证的门禁系统及其认证方法
WO2016172923A1 (zh) 视频检测方法、视频检测系统以及计算机程序产品
Gafurov et al. User authentication based on foot motion
Fernandez-Lopez et al. Optimizing resources on smartphone gait recognition
Haque et al. Heartbeat signal from facial video for biometric recognition
Li et al. Sensing in-air signature motions using smartwatch: A high-precision approach of behavioral authentication
Sikarwar Finger vein recognition using local directional pattern
Bhargava et al. Fingerprint matching of normalized image based on Euclidean distance
JP5373128B2 (ja) 人物特定装置、人物特定方法及び人物特定プログラム
Garg et al. Biometric authentication using soft biometric traits
Burgbacher et al. A behavioral biometric challenge and response approach to user authentication on smartphones
Samà et al. Gait identification by means of box approximation geometry of reconstructed attractors in latent space

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: 17934178

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17934178

Country of ref document: EP

Kind code of ref document: A1