CN105468951A - Method and device for identity recognition through electrocardiographic feature and wearable device - Google Patents
Method and device for identity recognition through electrocardiographic feature and wearable device Download PDFInfo
- Publication number
- CN105468951A CN105468951A CN201510796544.8A CN201510796544A CN105468951A CN 105468951 A CN105468951 A CN 105468951A CN 201510796544 A CN201510796544 A CN 201510796544A CN 105468951 A CN105468951 A CN 105468951A
- Authority
- CN
- China
- Prior art keywords
- original electro
- cardiologic signals
- ripple
- cardiologic
- proper vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/15—Biometric patterns based on physiological signals, e.g. heartbeat, blood flow
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention provides a method and device for identity recognition through an electrocardiographic feature and a wearable device. The method comprises the following steps: acquiring an original electrocardiosignal of a user by an electrocardiography transducer; determining a feature vector corresponding to the original electrocardiosignal, wherein the feature vector comprises time domain feature data of the original electrocardiosignal and frequency domain feature data of the original electrocardiosignal; and determining a user identity corresponding to the original electrocardiosignal through a trained large distance nearest neighbor algorithm corresponding to the feature vector and the original electrocardiosignal. The technical scheme of the invention can greatly improve the accuracy and safety of user identity authentication.
Description
Technical field
The application relates to wearable device technical field, particularly relates to a kind of method and device, the wearable device that are carried out identification by ecg characteristics.
Background technology
Along with development and the growth in the living standard of society, and the fast development of mobile Internet, traditional identification mode more and more can not meet people to the demand of authentication at secure context.Biometrics identification technology is in order to solve the problems referred to above, by recognition methodss such as fingerprint, face, voice, irises, certification is carried out to user identity, and be widely used at smart machine and mobile field, but, above-mentioned biological characteristic, owing to can be replicated and record, is therefore easily easily forged.
Summary of the invention
In view of this, the application provides a kind of new technical scheme, can solve the technical matters existed in above-mentioned prior art.
For achieving the above object, the application provides technical scheme as follows:
According to the first aspect of the application, propose a kind of method of being carried out identification by ecg characteristics, comprising:
The original electro-cardiologic signals of user is gathered by EGC sensor;
Determine the proper vector of described original electro-cardiologic signals, described proper vector comprises the frequency domain character data of temporal signatures data corresponding to described original electro-cardiologic signals and described original electro-cardiologic signals;
The large nearest adjacent algorithm of having trained corresponding with described original electro-cardiologic signals by described proper vector determines the user identity that described original electro-cardiologic signals is corresponding.
According to the second aspect of the application, propose a kind of device being carried out identification by ecg characteristics, comprising:
Signal acquisition module, for gathering the original electro-cardiologic signals of user by EGC sensor;
First determination module, for determining the described original electro-cardiologic signals characteristic of correspondence vector that described signal acquisition module gathers, described proper vector comprises the temporal signatures data of described original electro-cardiologic signals and the frequency domain character data of described original electro-cardiologic signals;
Second determination module, the large nearest adjacent algorithm of having trained corresponding with described original electro-cardiologic signals by the described proper vector determined by described first determination module determines the user identity that described original electro-cardiologic signals is corresponding.
According to the third aspect of the application, propose a kind of wearable device, described wearable device comprises:
Processor; For storing the storer of described processor executable;
Wherein, described processor, for gathering the original electro-cardiologic signals of user by EGC sensor;
Determine described original electro-cardiologic signals characteristic of correspondence vector, described proper vector comprises the temporal signatures data of described original electro-cardiologic signals and the frequency domain character data of described original electro-cardiologic signals;
The large nearest adjacent algorithm of having trained corresponding with described original electro-cardiologic signals by described proper vector determines the user identity that described original electro-cardiologic signals is corresponding.
From above technical scheme, the application gathers the original electro-cardiologic signals of user by EGC sensor, determine the proper vector of original electro-cardiologic signals, by the user identity that the large nearest adjacent algorithm determination original electro-cardiologic signals of having trained that proper vector is corresponding with original electro-cardiologic signals is corresponding, because proper vector comprises temporal signatures data and the frequency domain character data of original electro-cardiologic signals, the matrix model used in the large nearest adjacent algorithm of having trained can be obtained by the method for machine learning, the target of machine learning is as the criterion with the accuracy rate of the sorting algorithm defined by large-spacing minimum distance height of trying one's best, therefore by ECG, authentication is carried out to user, greatly can improve accuracy and the security of authenticating user identification.
Accompanying drawing explanation
Figure 1A shows the schematic flow sheet being carried out the method for identification by ecg characteristics according to an exemplary embodiment of the present invention;
Figure 1B shows the schematic diagram of the original electro-cardiologic signals according to an exemplary embodiment of the present invention;
Fig. 2 A shows the schematic flow sheet of the spectral feature data of the determination original electro-cardiologic signals according to an exemplary embodiment of the present invention;
Fig. 2 B shows the schematic diagram by the electrocardiosignal after wavelet transformation filtering noise according to an exemplary embodiment of the present invention;
Fig. 3 A shows the schematic flow sheet of the temporal signatures data of the determination original electro-cardiologic signals according to an exemplary embodiment of the present invention;
Fig. 3 B shows the sequential of electrocardiosignal and the schematic diagram of amplitude Characteristics;
Fig. 3 C shows the schematic diagram of the hardware circuit detection R ripple that Fig. 3 A adopts;
Fig. 3 D shows the circuit diagram for detecting dynamic threshold in Fig. 3 C;
Fig. 4 A shows the schematic flow sheet being carried out the method for identification by ecg characteristics in accordance with a further exemplary embodiment of the present invention;
Fig. 4 B shows the schematic diagram of electrocardiosignal corresponding to the reference feature vector of two;
Fig. 5 shows the structural representation of the wearable device according to an exemplary embodiment of the present invention;
Fig. 6 shows the structural representation being carried out the device of identification by ecg characteristics according to an exemplary embodiment of the present invention;
Fig. 7 shows the structural representation being carried out the device of identification by ecg characteristics in accordance with a further exemplary embodiment of the present invention.
Embodiment
Here will be described exemplary embodiment in detail, its sample table shows in the accompanying drawings.When description below relates to accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawing represents same or analogous key element.Embodiment described in following exemplary embodiment does not represent all embodiments consistent with the application.On the contrary, they only with as in appended claims describe in detail, the example of apparatus and method that some aspects of the application are consistent.
Only for describing the object of specific embodiment at term used in this application, and not intended to be limiting the application." one ", " described " and " being somebody's turn to do " of the singulative used in the application and appended claims is also intended to comprise most form, unless context clearly represents other implications.It is also understood that term "and/or" used herein refer to and comprise one or more project of listing be associated any or all may combine.
Term first, second, third, etc. may be adopted although should be appreciated that to describe various information in the application, these information should not be limited to these terms.These terms are only used for the information of same type to be distinguished from each other out.Such as, when not departing from the application's scope, the first information also can be called as the second information, and similarly, the second information also can be called as the first information.Depend on linguistic context, word as used in this " if " can be construed as into " ... time " or " when ... time " or " in response to determining ".
The features such as compared with biological characteristic of the prior art, the electrocardiosignal (ECG) of human body is determined by the cardiac structure of each individuality, and it has universality, uniqueness, easily gathers, permanent, and, ECG also has and is only present in live body, is not easily imitated, not easily the advantage such as loss.Therefore, the biometrics identification technology based on ECG has very important effect for the security improving authentication.
The application gathers the original electro-cardiologic signals of user by EGC sensor, determine the proper vector of original electro-cardiologic signals, by the user identity that the large nearest adjacent algorithm model determination original electro-cardiologic signals of having trained that proper vector is corresponding with original electro-cardiologic signals is corresponding, because proper vector comprises temporal signatures data and the frequency domain character data of original electro-cardiologic signals, therefore by ECG, authentication is carried out to user, greatly can improve accuracy and the security of authenticating user identification.
For being further described the application, provide the following example:
Figure 1A shows the schematic flow sheet being carried out the method for identification by ecg characteristics according to an exemplary embodiment of the present invention, and Figure 1B shows the schematic diagram of the original electro-cardiologic signals according to an exemplary embodiment of the present invention; The present embodiment can be applied on wearable device, such as, on the equipment such as Intelligent bracelet, Intelligent bracelet can be provided with EGC sensor, as shown in Figure 1A, comprise the steps:
Step 101, gathers the original electro-cardiologic signals of user by EGC sensor.
Step 102, determine original electro-cardiologic signals characteristic of correspondence vector, proper vector comprises the temporal signatures data of original electro-cardiologic signals and the frequency domain character data of original electro-cardiologic signals.
Step 103, by the user identity that the large nearest adjacent algorithm determination original electro-cardiologic signals of having trained that proper vector is corresponding with original electro-cardiologic signals is corresponding.
In a step 101, as shown in Figure 1B, original electro-cardiologic signals has stronger noise, and can As time goes on change, but the QRS wave group of the electrocardiosignal that do not gather in the same time of same user, P ripple, T ripple are substantially identical.
In a step 102, in one embodiment, the frequency domain character data of original electro-cardiologic signals can comprise wavelet conversion coefficient corresponding to original electro-cardiologic signals, auto-correlation and discrete cosine transform coefficient, fourier transform coefficient, HHT (Hilbert-Hwang) conversion coefficient etc., and the application does not limit the conversion of concrete frequency domain.
In step 103, in one embodiment, can by carrying out the electrocardiogram (ECG) data of existing subscriber training the large nearest adjacent algorithm model obtaining and trained.In one embodiment, can complete under line the training of large nearest adjacent algorithm, in the model parameter needing directly to use training to obtain when realizing authentication by the application.The matrix model used in the large nearest adjacent algorithm model of having trained can be obtained by the method for machine learning, the target of machine learning is as the criterion with the accuracy rate of the sorting algorithm defined by large-spacing minimum distance height of trying one's best, thus can guarantee the accuracy of identification.
Seen from the above description, the embodiment of the present invention gathers the original electro-cardiologic signals of user by EGC sensor, determine the proper vector of original electro-cardiologic signals, by the user identity that the large nearest adjacent algorithm determination original electro-cardiologic signals of having trained that proper vector is corresponding with original electro-cardiologic signals is corresponding, because proper vector comprises temporal signatures data and the frequency domain character data of original electro-cardiologic signals, the matrix model used in the large nearest adjacent algorithm of having trained can be obtained by the method for machine learning, the target of machine learning is as the criterion with the accuracy rate of the sorting algorithm defined by large-spacing minimum distance height of trying one's best, therefore by ECG, authentication is carried out to user, greatly can improve accuracy and the security of authenticating user identification.
Fig. 2 A shows the schematic flow sheet of the spectral feature data of the determination original electro-cardiologic signals according to an exemplary embodiment of the present invention, and Fig. 2 B shows the schematic diagram by the electrocardiosignal after wavelet transformation filtering noise according to an exemplary embodiment of the present invention; As Fig. 2 A institute, comprise the steps:
Step 201, carries out wavelet transformation to original electro-cardiologic signals, obtains the wavelet coefficient of original electro-cardiologic signals.
Step 202, is defined as the frequency domain character data of original electro-cardiologic signals by wavelet coefficient.
Step 203, carries out auto-correlation and discrete cosine transform to the electrocardiosignal after wavelet transformation, obtains the auto-correlation after auto-correlation and discrete cosine transform and discrete cosine transform coefficient.
Step 204, is defined as the frequency domain character data of original electro-cardiologic signals by auto-correlation and discrete cosine transform coefficient.
In step 201 and step 202, can pass through wavelet transformation can by the signal decomposition of different frequency in original electro-cardiologic signals out, due to the high frequency noise of reflection original electro-cardiologic signals main on low yardstick, the low-frequency noise of main reflection original electro-cardiologic signals on high yardstick, therefore the medium scale after the application adopts wavelet transformation is analyzed original electro-cardiologic signals, thus can effectively useful signal and undesired signal be distinguished.In one embodiment, can carry out wavelet decomposition by one group of coefficient high pass undetermined and low-pass filter to original electro-cardiologic signals, thus obtain wavelet coefficient corresponding to original electro-cardiologic signals, wavelet coefficient can comprise one-level scale coefficient and wavelet coefficient.In one embodiment, can be undertaken being shifted by the hardware platform of FPGA and wavelet transformation that the mode of addition realizes original electro-cardiologic signals, and the logic of displacement and additive operation simply, is easy to realization.As shown in Figure 2 B, by wavelet transformation to after original electro-cardiologic signals filtering noise, the noise of original electro-cardiologic signals is effectively removed, and electrocardiosignal is more regular.
In step 204, in one embodiment, auto-correlation computation can be carried out to filtered electrocardiosignal, can be eliminated in electrocardiosignal identifying irrelevant signal section by auto-correlation computation, again the signal after auto-correlation computation is carried out discrete cosine transform, and then obtain auto-correlation and discrete cosine transform coefficient.In another embodiment, Fourier transform or HHT (Hilbert-Hwang) conversion etc. can also be carried out, using the frequency domain character data of the coefficient after conversion as electrocardiosignal to filtered electrocardiosignal.
The present embodiment by wavelet transformation to original electro-cardiologic signals filtering noise, the noise of original electro-cardiologic signals can be made effectively to be removed, make electrocardiosignal more regular, thus guarantee that the wavelet coefficient of original electro-cardiologic signals on each yardstick and auto-correlation and discrete cosine transform coefficient more can represent the feature of original electro-cardiologic signals at frequency domain exactly as frequency domain character data.
Fig. 3 A shows the schematic flow sheet of the temporal signatures data of the determination original electro-cardiologic signals according to an exemplary embodiment of the present invention, Fig. 3 B shows the sequential of electrocardiosignal and the schematic diagram of amplitude Characteristics, Fig. 3 C shows the schematic diagram of the hardware circuit detection R ripple that Fig. 3 A adopts, and Fig. 3 D shows the circuit diagram for detecting dynamic threshold in Fig. 3 C; As shown in Figure 3A, comprise the steps:
Step 301, passes through the first comparer point-by-point comparison by the wavelet coefficient on each yardstick and predetermined threshold value.
Step 302, when the first comparer detects the wavelet coefficient being less than predetermined threshold value, the value storage remembered by counter is in register.
Step 303, the numerical value continuing to count to get under the effect of counter in clock signal is within distance detects the time period that the minimizing numerical value difference QRS ripple of original electro-cardiologic signals is corresponding, detect that the wavelet coefficient obtained in register reaches maximum value, determine the R crest detected in original electro-cardiologic signals.
Step 304, according to the P ripple in the center extraction original electro-cardiologic signals of R crest and T ripple.
Step 305, according to the temporal signatures data of R crest, P ripple, T ripple determination original electro-cardiologic signals, temporal signatures data comprise: the range value of the crest location of the crest location of R ripple, the crest location of P ripple, T ripple, the range value of P ripple, R ripple, T ripple, the interval of the interval of P ripple and R crest, T ripple and R crest, PR section, ST section.
From the electrocardiosignal shown in Fig. 3 B, by carrying out the wavelet transform based on quadratic spline to original electro-cardiologic signals, known about the research of wavelet transformation through people such as Mallat, the singular point of original electro-cardiologic signals is if the intersection point of a pair rising edge and negative edge, then the signal that this intersection point is corresponding becomes the zero point of a negative maximum and positive maximum after wavelet transformation.And the position that the R ripple of electrocardiosignal in the application occurs is just in time the zero crossing position that each yardstick extreme value is right, therefore the application's right zero crossing that only needs to detect extreme value on the wavelet transformation of original electro-cardiologic signals on each yardstick just can detect the crest location of R ripple.P ripple and T ripple also can be extracted by identical method, and the temporal signatures data of electrocardiosignal can shown in reference diagram 3B.In one embodiment, P ripple and T ripple can be detected in conjunction with the testing result of previous R ripple, because the waveform of a lot of low-frequency noise and base drift and P ripple is just the same, the present embodiment is by detecting P ripple and T ripple in conjunction with the testing result of R ripple, such as, after the crest location determining R ripple, within the front and back a period of time centered by the crest location of this R ripple, find the position of P ripple forward (such as, in the time period of the crest location-150ms of the crest location-250ms to R of R), and at [crest location+the 170ms of R, crest location+the 400ms of R] time period in detect T ripple, because PQRST ripple is continuous print, with the position of R ripple for reference detects Q forwards, backwards respectively, S ripple, be benchmaring P ripple again with Q, with S ripple for benchmaring T ripple, thus the speed of detection can be improved, and reduce the error rate detected.In one embodiment, can also the characteristic in time domain needs in the application using Q ripple and S ripple temporal signatures data, thus electrocardiosignal character representation in the time domain can be improved further.
In one embodiment, the detection of the temporal signatures data of electrocardiosignal can be realized by the mode of hardware circuit, as shown in Figure 3 C, exemplary illustration is carried out for the crest location being detected R ripple by hardware circuit, wavelet coefficient h on each yardstick after wavelet transformation is carried out point-by-point comparison with the predetermined threshold value be stored in the first register 31 by the first comparer 32, first comparer 32 according to the comparative result of the wavelet coefficient h on each yardstick and predetermined threshold value for the second register 33 provides logical signal, with the count value making the second register 33 memory counter 34 obtain, first counter 34 continues counting under the effect of clock clk1, when in distance, the numerical value of the first counter 34 detects that the minimizing numerical value of wavelet coefficient differs the time of a R wave width (such as, a maximum value is detected again 0.1s), now the first comparer 32 and the second comparer 35 all can provide the signal of logical one to door 36, thus determine a R crest to be detected, record the center of R crest and the width of R ripple that detect.
In one embodiment, can to adopt based on the center of R crest and above-mentioned similar method extracts P ripple in original electro-cardiologic signals and T ripple, and then obtain the temporal signatures data described in above-mentioned steps 305.
The factor such as fluctuation, base drift due to electrocardiosignal makes QRS ripple amplitude in the same time can be not different yet, and then make on synchronization different scale that maximum value is in the same time not identical with minimal value with same yardstick, therefore the application can adopt the threshold value that dynamic method detection temporal signatures data use, and can be improved the accuracy extracting temporal signatures data by dynamic threshold.As shown in Figure 3 D, value in 3rd register 38 can be predisposed to 0, wavelet coefficient h pointwise on each yardstick and the value be deposited with in the 3rd register 38 compare by the 3rd comparer 37, if the 3rd comparer 37 finds larger wavelet coefficient, then send a logic control signal to the 3rd register 38, the threshold value that the range value making the 3rd register 38 larger according to this according to this logic control signal calculates as follows:
Threshold value for corresponding during positive maximum: MAX==p (a*max1+b*max2);
Threshold value for corresponding during negative minimal value: MIN==q (a*min1+b*min2).
Wherein, max1 is the maximal value of the wavelet coefficient detected in the sense cycle of first dynamic threshold, min1 is the wavelet coefficient minimum value detected in the sense cycle of first dynamic threshold, max2 is the maximal value of the wavelet coefficient detected in the sense cycle of second dynamic threshold, min2 is the minimum value of the wavelet coefficient detected in the sense cycle of second dynamic threshold, a+b=1, represents each self-corresponding weight, p and q be less than 1 positive number.
The above-mentioned threshold value calculated is stored in the first register 31, second counter 34 controls current count when reaching the sense cycle of a dynamic threshold, threshold value in first register 33 is reset, thus prepares the renewal of the threshold value in the sense cycle of next dynamic threshold.
In the present embodiment, achieved the detection of temporal signatures data by the hardware mode such as register, comparer, improve the real-time detecting temporal signatures data, threshold value is different along with the difference of electrocardiosignal, thus can improve the precision of temporal signatures data; In addition, the speed of detection can be improved by detecting P ripple and T ripple in conjunction with the testing result of R ripple, and reduce the error rate detected.
Fig. 4 A shows the schematic flow sheet being carried out the method for identification by ecg characteristics in accordance with a further exemplary embodiment of the present invention, and Fig. 4 B shows the schematic diagram of electrocardiosignal corresponding to the reference feature vector of two; As shown in Figure 4 A, comprise the steps:
Step 401, gathers the original electro-cardiologic signals of user by EGC sensor.
Step 402, determine original electro-cardiologic signals characteristic of correspondence vector, proper vector comprises the temporal signatures data of original electro-cardiologic signals and the frequency domain character data of original electro-cardiologic signals.
Step 403, calculates at least one Similarity value between proper vector and at least one reference feature vector stored according to the large nearest adjacent algorithm of having trained.
Step 404, determines the user ID that maximum Similarity value at least one Similarity value is corresponding.
Step 405, is identified as user corresponding to original electro-cardiologic signals by user ID corresponding for maximum Similarity value.
The description of step 401 and step 402 see the associated description shown in above-mentioned Figure 1A, can be not described in detail in this.
In step 403, proper vector is being obtained by above-mentioned steps
after, Similarity value can be obtained by following formulae discovery:
Wherein,
represent the proper vector of the user of original electro-cardiologic signals,
for r the reference feature vector stored, r is positive integer, and M is the matrix model obtained by the method for machine learning, and T represents the transposition of vector, the weight coefficient that described in the element representation in described matrix model, proper vector is corresponding.
Take wearable device as Intelligent bracelet for example carries out exemplary illustration, if stored user A and user B in Intelligent bracelet about reference feature vector corresponding to respective electrocardiosignal
with
as shown in Figure 4 B, user A is not identical in the shape of time domain with the electrocardiosignal of user B, and therefore respective reference feature vector also can not be identical, and the present embodiment can calculate proper vector by large nearest adjacent algorithm
respectively with reference feature vector
with
between Similarity value be d
1and d
2, by from d
1and d
2the identify label of the user finding higher value to be collected by EGC sensor as Intelligent bracelet, such as, d
1be greater than d
2, then can by proper vector
be identified as user A.
In one embodiment, same user can corresponding multiple reference feature vector, multiple reference feature vector can by user move and static time reference feature vector, such as, by the proper vector obtained above by the electrocardiosignal collected
the multiple feature reference vectors corresponding with same user adopt said method to calculate multiple Similarity value, find Similarity value larger in multiple Similarity value, namely may identify this user and be in state corresponding to the Similarity value larger with this, such as, motion state or stationary state etc.
In the present embodiment, due to can online by the method for learning distance metric (metriclearning) under training carried out to large nearest adjacent algorithm obtain matrix model, the weight coefficient that described in element representation in matrix model, proper vector is corresponding, therefore the present invention is more conducive to being realized by the mode of hardware, thus can solve the problem of software algorithm longer and degree of accuracy deficiency consuming time in implementation procedure; In addition, the method in conjunction with machine learning obtains the discrimination that matrix model can improve electrocardiosignal.
Fig. 5 shows the structural representation of the wearable device according to an exemplary embodiment of the present invention; As shown in Figure 5, EGC sensor 51 collects original electro-cardiologic signals, signal processing module 521 pairs of original electro-cardiologic signals carry out wavelet transformation and then carry out filtering process to original electro-cardiologic signals, obtain the coefficient of wavelet transformation, characteristic vector pickup module 522 obtains temporal signatures data and frequency domain character data according to said method embodiment, the Similarity value of the reference feature vector that model 523 is stored by the large nearest adjacent algorithm calculating proper vector of employing and memory module 53, result discrimination module 524 obtains identification result.Wherein, signal processing module 521, characteristic vector pickup module 522, model 523, result discrimination module 524 include in FPGA system 52.Memory module 53 also stores the matrix model that large nearest adjacent algorithm uses, this matrix model is obtained by the mode of training under line, thus reduce the computation complexity of FPGA system 52, shorten the time of electrocardiosignal identification, improve the efficiency of identification.
Fig. 6 shows the structural representation being carried out the device of identification by ecg characteristics according to an exemplary embodiment of the present invention; As shown in Figure 6, this device carrying out identification by ecg characteristics can comprise: signal acquisition module 61, first determination module 62, second determination module 63.Wherein:
Signal acquisition module 61, for gathering the original electro-cardiologic signals of user by EGC sensor;
First determination module 62, for determining the original electro-cardiologic signals characteristic of correspondence vector that signal acquisition module 61 gathers, proper vector comprises the temporal signatures data of original electro-cardiologic signals and the frequency domain character data of original electro-cardiologic signals;
Second determination module 63, with the user identity that the large nearest adjacent algorithm determination original electro-cardiologic signals of having trained that the proper vector determined by the first determination module 62 is corresponding with original electro-cardiologic signals is corresponding.
Fig. 7 shows the structural representation being carried out the device of identification by ecg characteristics in accordance with a further exemplary embodiment of the present invention; As shown in Figure 7, on above-mentioned basis embodiment illustrated in fig. 6, in one embodiment, the first determination module 62 can comprise:
Wavelet transform unit 621, for carrying out wavelet transformation to original electro-cardiologic signals, obtains the wavelet coefficient of original electro-cardiologic signals on each yardstick;
First determining unit 622, the wavelet coefficient for wavelet transform unit 621 being obtained is defined as the frequency domain character data of original electro-cardiologic signals.
In one embodiment, the first determination module 62 can comprise:
First arithmetic element 623, for carrying out auto-correlation computation and discrete cosine transform to the electrocardiosignal after wavelet transform filtering, obtains the auto-correlation after auto-correlation and discrete cosine transform and discrete cosine transform coefficient;
Second determining unit 624, is defined as the frequency domain character data of original electro-cardiologic signals for the auto-correlation that the first arithmetic element 623 computing obtained and discrete cosine transform coefficient.
In one embodiment, the first determination module 62 can comprise:
Comparing unit 625, the wavelet coefficient on each yardstick wavelet transform unit obtained and predetermined threshold value are by the first comparer point-by-point comparison;
Storage unit 626, for when comparing unit 625 represents that the wavelet coefficient that the first comparer detects is less than predetermined threshold value, the value storage remembered by counter is in register;
3rd determining unit 627, for the numerical value that continues to count to get under the effect of counter in clock signal within distance detects the time period that the minimizing numerical value difference QRS ripple of original electro-cardiologic signals is corresponding, detect that the range value obtained in register reaches maximum value, determine the R crest detected in original electro-cardiologic signals;
Feature extraction unit 628, P ripple in original electro-cardiologic signals and T ripple are extracted in the center of R crest for determining according to the 3rd determining unit 627;
4th determining unit 629, for the temporal signatures data of the described R crest determined according to the 3rd determining unit 627, P ripple that feature extraction unit is extracted, T ripple determination original electro-cardiologic signals, temporal signatures data comprise: the crest location of the crest location of R ripple, the crest location of P ripple, T ripple, the range value of P ripple, the range value of R ripple, the range value of T ripple, the interval of the crest location of the interval of the crest location of P and the crest location of R ripple, the crest location of T ripple and R ripple, PR section, ST section.
In one embodiment, the second determination module 63 can comprise:
Computing unit 631, for calculating at least one Similarity value between proper vector and at least one reference feature vector stored that the first determination module determines according to the large nearest adjacent algorithm of having trained;
5th determining unit 632, for determining the user ID that maximum Similarity value at least one Similarity value that computing unit 631 calculates is corresponding;
Recognition unit 633, the user ID corresponding for the maximum Similarity value in the 5th determining unit 632 being obtained is identified as user corresponding to original electro-cardiologic signals.
In one embodiment, computing unit 631 is by least one Similarity value of following formulae discovery:
Wherein,
represent the proper vector of the user of original electro-cardiologic signals,
for r the reference feature vector stored, r is positive integer, and M is the matrix model obtained by the device of machine learning, the weight coefficient that the element representation proper vector in matrix model is corresponding.
Above-described embodiment is visible, the application gathers the original electro-cardiologic signals of user by EGC sensor, determine the proper vector of original electro-cardiologic signals, by the user identity that the large nearest adjacent algorithm model determination original electro-cardiologic signals of having trained that proper vector is corresponding with original electro-cardiologic signals is corresponding, because proper vector comprises temporal signatures data and the frequency domain character data of original electro-cardiologic signals, therefore by ECG, authentication is carried out to user, greatly can improve accuracy and the security of authenticating user identification.
Those skilled in the art, at consideration instructions and after putting into practice invention disclosed herein, will easily expect other embodiment of the application.The application is intended to contain any modification of the application, purposes or adaptations, and these modification, purposes or adaptations are followed the general principle of the application and comprised the undocumented common practise in the art of the application or conventional techniques means.Instructions and embodiment are only regarded as exemplary, and true scope and the spirit of the application are pointed out by claim below.
Also it should be noted that, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, commodity or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, commodity or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, commodity or the equipment comprising described key element and also there is other identical element.
The foregoing is only the preferred embodiment of the application, not in order to limit the application, within all spirit in the application and principle, any amendment made, equivalent replacements, improvement etc., all should be included within scope that the application protects.
Claims (13)
1. carried out a method for identification by ecg characteristics, it is characterized in that, described method comprises:
The original electro-cardiologic signals of user is gathered by EGC sensor;
Determine described original electro-cardiologic signals characteristic of correspondence vector, described proper vector comprises the temporal signatures data of described original electro-cardiologic signals and the frequency domain character data of described original electro-cardiologic signals;
The large nearest adjacent algorithm of having trained corresponding with described original electro-cardiologic signals by described proper vector determines the user identity that described original electro-cardiologic signals is corresponding.
2. method according to claim 1, is characterized in that, the described proper vector determining described original electro-cardiologic signals, comprising:
Wavelet transformation is carried out to described original electro-cardiologic signals, obtains the wavelet coefficient of described original electro-cardiologic signals on each yardstick;
Described wavelet coefficient is defined as the frequency domain character data of described original electro-cardiologic signals.
3. method according to claim 2, is characterized in that, the described proper vector determining described original electro-cardiologic signals, comprising:
Auto-correlation computation is carried out to the electrocardiosignal after described wavelet transform filtering, and discrete cosine transform, obtain the auto-correlation after described auto-correlation and discrete cosine transform and discrete cosine transform coefficient;
Described auto-correlation and discrete cosine transform coefficient are defined as the frequency domain character data of described original electro-cardiologic signals.
4. method according to claim 2, is characterized in that, the described proper vector determining described original electro-cardiologic signals, comprising:
Wavelet coefficient on described each yardstick and predetermined threshold value are passed through the first comparer point-by-point comparison;
When described first comparer detects the wavelet coefficient being less than described predetermined threshold value, the value storage remembered by counter is in register;
The numerical value continuing to count to get under the effect of described counter in clock signal is within distance detects the time period that the minimizing numerical value difference QRS ripple of described original electro-cardiologic signals is corresponding, detect that the range value obtained in described register reaches maximum value, determine the R crest detected in described original electro-cardiologic signals;
P ripple in described original electro-cardiologic signals and T ripple is extracted according to the center of described R crest;
The temporal signatures data of described original electro-cardiologic signals are determined according to described R crest, described P ripple, described T ripple, described temporal signatures data comprise: the crest location of described R ripple, the crest location of described P ripple, the crest location of described T ripple, the range value of described P ripple, the range value of described R ripple, the range value of described T ripple, the interval of the crest location of the interval of the crest location of described P and the crest location of described R ripple, the crest location of described T ripple and described R ripple, PR section, ST section.
5., according to the arbitrary described method of claim 1-4, it is characterized in that, the described large nearest adjacent algorithm of having trained corresponding with described original electro-cardiologic signals by described proper vector is determined to comprise the user identity that described original electro-cardiologic signals is corresponding:
At least one Similarity value between described proper vector and the reference feature vector of at least one stored is calculated according to the large nearest adjacent algorithm of having trained;
The user ID that maximum Similarity value at least one Similarity value described in determining is corresponding;
User ID corresponding for described maximum Similarity value is identified as user corresponding to described original electro-cardiologic signals.
6. method according to claim 5, it is characterized in that, describedly face most at least one Similarity value that distance algorithm calculates between described proper vector and at least one reference feature vector stored, by least one Similarity value described in following formulae discovery according to large-spacing:
Wherein,
represent the proper vector of the user of original electro-cardiologic signals,
for r the reference feature vector stored, r is positive integer, and M is the matrix model obtained by the method for machine learning, the weight coefficient that described in the element representation in described matrix model, proper vector is corresponding.
7. carried out a device for identification by ecg characteristics, it is characterized in that, described device comprises:
Signal acquisition module, for gathering the original electro-cardiologic signals of user by EGC sensor;
First determination module, for determining the described original electro-cardiologic signals characteristic of correspondence vector that described signal acquisition module gathers, described proper vector comprises the temporal signatures data of described original electro-cardiologic signals and the frequency domain character data of described original electro-cardiologic signals;
Second determination module, the large nearest adjacent algorithm of having trained corresponding with described original electro-cardiologic signals by the described proper vector determined by described first determination module determines the user identity that described original electro-cardiologic signals is corresponding.
8. device according to claim 7, is characterized in that, described first determination module comprises:
Wavelet transform unit, for carrying out wavelet transformation to described original electro-cardiologic signals, obtains the wavelet coefficient of described original electro-cardiologic signals on each yardstick;
First determining unit, the described wavelet coefficient for described wavelet transform unit being obtained is defined as the frequency domain character data of described original electro-cardiologic signals.
9. device according to claim 8, is characterized in that, described first determination module comprises:
First arithmetic element, for carrying out auto-correlation computation and discrete cosine transform to the electrocardiosignal after described wavelet transform filtering, obtains the auto-correlation after described auto-correlation and discrete cosine transform and discrete cosine transform coefficient;
Second determining unit, is defined as the frequency domain character data of described original electro-cardiologic signals for the described auto-correlation that described first arithmetic element computing obtained and discrete cosine transform coefficient.
10. device according to claim 8, is characterized in that, described first determination module comprises:
Comparing unit, the wavelet coefficient on the described each yardstick described wavelet transform unit obtained and predetermined threshold value are by the first comparer point-by-point comparison;
Storage unit, during for representing that described first comparer detects the wavelet coefficient being less than described predetermined threshold value at described comparing unit, the value storage remembered by counter is in register;
3rd determining unit, for the numerical value that continues to count to get under the effect of described counter in clock signal within distance detects the time period that the minimizing numerical value difference QRS ripple of described original electro-cardiologic signals is corresponding, detect that the range value obtained in described register reaches maximum value, determine the R crest detected in described original electro-cardiologic signals;
Feature extraction unit, P ripple in described original electro-cardiologic signals and T ripple are extracted in the center for the described R crest determined according to described 3rd determining unit;
4th determining unit, for the described R crest determined according to described 3rd determining unit, the described P ripple that described feature extraction unit is extracted, described T ripple determines the temporal signatures data of described original electro-cardiologic signals, described temporal signatures data comprise: the crest location of described R ripple, the crest location of described P ripple, the crest location of described T ripple, the range value of described P ripple, the range value of described R ripple, the range value of described T ripple, the interval of the crest location of described P and the crest location of described R ripple, the interval of the crest location of described T ripple and the crest location of described R ripple, PR section, ST section.
11. according to the arbitrary described device of claim 7-10, and it is characterized in that, described second determination module comprises:
Computing unit, for calculating at least one Similarity value between described proper vector and at least one reference feature vector stored that described first determination module determines according to the large nearest adjacent algorithm of having trained;
5th determining unit, for determine described computing unit calculate described in user ID corresponding to maximum Similarity value at least one Similarity value;
Recognition unit, the user ID corresponding for the maximum Similarity value in described 5th determining unit being obtained is identified as user corresponding to described original electro-cardiologic signals.
12. devices according to claim 11, is characterized in that, described computing unit is by least one Similarity value described in following formulae discovery:
Wherein,
represent the proper vector of the user of original electro-cardiologic signals,
for r the reference feature vector stored, r is positive integer, and M is the matrix model obtained by the device of machine learning, the weight coefficient that described in the element representation in described matrix model, proper vector is corresponding.
13. 1 kinds of wearable devices, is characterized in that, described wearable device comprises:
Processor; For storing the storer of described processor executable;
Wherein, described processor, for gathering the original electro-cardiologic signals of user by EGC sensor;
Determine the proper vector of described original electro-cardiologic signals, described proper vector comprises the frequency domain character data of temporal signatures data corresponding to described original electro-cardiologic signals and described original electro-cardiologic signals;
The large nearest adjacent algorithm of having trained corresponding with described original electro-cardiologic signals by described proper vector determines the user identity that described original electro-cardiologic signals is corresponding.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510796544.8A CN105468951B (en) | 2015-11-17 | 2015-11-17 | Method and device, the wearable device of identification are carried out by ecg characteristics |
| PCT/CN2016/105720 WO2017084546A1 (en) | 2015-11-17 | 2016-11-14 | Wearable device-based user interest information determination method, device and wearable device |
| US15/584,911 US10163528B2 (en) | 2015-11-17 | 2017-05-02 | Determining user-interested information based on wearable device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510796544.8A CN105468951B (en) | 2015-11-17 | 2015-11-17 | Method and device, the wearable device of identification are carried out by ecg characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105468951A true CN105468951A (en) | 2016-04-06 |
| CN105468951B CN105468951B (en) | 2019-08-06 |
Family
ID=55606640
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510796544.8A Active CN105468951B (en) | 2015-11-17 | 2015-11-17 | Method and device, the wearable device of identification are carried out by ecg characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105468951B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106485213A (en) * | 2016-09-27 | 2017-03-08 | 鲁东大学 | A kind of utilization electrocardiosignal carries out the feature extracting method of automatic identification |
| WO2017084546A1 (en) * | 2015-11-17 | 2017-05-26 | 安徽华米信息科技有限公司 | Wearable device-based user interest information determination method, device and wearable device |
| CN106897693A (en) * | 2017-02-23 | 2017-06-27 | 深圳先进技术研究院 | A kind of personal identification method, apparatus and system |
| CN107239684A (en) * | 2017-05-22 | 2017-10-10 | 吉林大学 | A kind of feature learning method and system for ECG identifications |
| CN107374614A (en) * | 2017-07-25 | 2017-11-24 | 海南聚能科技创新研究院有限公司 | The automatic classification storage control method of electrocardiogram and system |
| GB2552035A (en) * | 2016-07-08 | 2018-01-10 | B-Secur Ltd | ECG Authentication method and apparatus |
| CN108304746A (en) * | 2017-01-11 | 2018-07-20 | 北京三星通信技术研究有限公司 | Method and apparatus of the update for the certification reference information of electrocardio authentication |
| WO2018152724A1 (en) * | 2017-02-23 | 2018-08-30 | 深圳先进技术研究院 | Identity recognition method, device and system |
| CN110869930A (en) * | 2019-10-08 | 2020-03-06 | 焦旭 | Identity recognition method based on cardiac shock signal, electronic device and storage medium |
| CN111063437A (en) * | 2019-12-12 | 2020-04-24 | 中科海微(北京)科技有限公司 | Personalized chronic disease analysis system |
| CN112257518A (en) * | 2020-09-30 | 2021-01-22 | 西安交通大学第二附属医院 | ECG identity recognition method based on WT and WOA-PNN algorithm |
| CN112401902A (en) * | 2020-12-04 | 2021-02-26 | 山东大学 | Electrocardio identity recognition method and system based on neural network time-frequency analysis combination |
| CN113221736A (en) * | 2021-05-11 | 2021-08-06 | 菏泽学院 | Electrocardio identity recognition method and system based on multi-feature collaborative non-negative matrix factorization |
| CN113440109A (en) * | 2021-08-02 | 2021-09-28 | 和也健康科技有限公司 | Human physiological parameter detection and monitoring system |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101773394A (en) * | 2010-01-06 | 2010-07-14 | 中国航天员科研训练中心 | Identification method and identification system using identification method |
| US20110150291A1 (en) * | 2009-12-22 | 2011-06-23 | Samsung Electronics Co., Ltd. | Method and apparatus for recognizing id in mobile terminal having a touchscreen |
| CN102274029A (en) * | 2011-05-25 | 2011-12-14 | 中国科学院深圳先进技术研究院 | Identity recognition method and system |
| CN103093133A (en) * | 2013-01-08 | 2013-05-08 | 西安电子科技大学 | Biological identity authentication method facing institute of electrical and electronic engineers (IEEE) 802.15.6 |
| CN103584852A (en) * | 2012-08-15 | 2014-02-19 | 深圳中科强华科技有限公司 | Personalized electrocardiogram intelligent auxiliary diagnosis device and method |
| CN103610457A (en) * | 2013-11-19 | 2014-03-05 | 深圳先进技术研究院 | Electrocardiosignal processing method and electrocardiosignal processing system |
| CN103714281A (en) * | 2013-12-12 | 2014-04-09 | 深圳先进技术研究院 | Identity recognition method based on electrocardiosignals |
| CN104055499A (en) * | 2014-06-16 | 2014-09-24 | 朱宇东 | Wearable intelligent hand ring and method for continuously monitoring human body physiological signs |
| CN104063645A (en) * | 2014-07-01 | 2014-09-24 | 清华大学深圳研究生院 | Identity recognition method based on ECG dynamic self-updating samples |
| CN104382584A (en) * | 2014-11-24 | 2015-03-04 | 北京联合大学 | Wearable electrocardiogram signal monitoring and evaluation system and implementation method thereof |
| CN104382590A (en) * | 2014-12-11 | 2015-03-04 | 赖大坤 | Automatic shockable rhythm identification and classification method combined with electrocardio time-frequency domain feature analysis |
| CN104510463A (en) * | 2014-12-08 | 2015-04-15 | 华南理工大学 | Wearable device based electroencephalograph detection device |
| CN104573458A (en) * | 2014-12-30 | 2015-04-29 | 深圳先进技术研究院 | Identity recognition method, device and system based on electrocardiogram signals |
-
2015
- 2015-11-17 CN CN201510796544.8A patent/CN105468951B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110150291A1 (en) * | 2009-12-22 | 2011-06-23 | Samsung Electronics Co., Ltd. | Method and apparatus for recognizing id in mobile terminal having a touchscreen |
| CN101773394A (en) * | 2010-01-06 | 2010-07-14 | 中国航天员科研训练中心 | Identification method and identification system using identification method |
| CN102274029A (en) * | 2011-05-25 | 2011-12-14 | 中国科学院深圳先进技术研究院 | Identity recognition method and system |
| CN103584852A (en) * | 2012-08-15 | 2014-02-19 | 深圳中科强华科技有限公司 | Personalized electrocardiogram intelligent auxiliary diagnosis device and method |
| CN103093133A (en) * | 2013-01-08 | 2013-05-08 | 西安电子科技大学 | Biological identity authentication method facing institute of electrical and electronic engineers (IEEE) 802.15.6 |
| CN103610457A (en) * | 2013-11-19 | 2014-03-05 | 深圳先进技术研究院 | Electrocardiosignal processing method and electrocardiosignal processing system |
| CN103714281A (en) * | 2013-12-12 | 2014-04-09 | 深圳先进技术研究院 | Identity recognition method based on electrocardiosignals |
| CN104055499A (en) * | 2014-06-16 | 2014-09-24 | 朱宇东 | Wearable intelligent hand ring and method for continuously monitoring human body physiological signs |
| CN104063645A (en) * | 2014-07-01 | 2014-09-24 | 清华大学深圳研究生院 | Identity recognition method based on ECG dynamic self-updating samples |
| CN104382584A (en) * | 2014-11-24 | 2015-03-04 | 北京联合大学 | Wearable electrocardiogram signal monitoring and evaluation system and implementation method thereof |
| CN104510463A (en) * | 2014-12-08 | 2015-04-15 | 华南理工大学 | Wearable device based electroencephalograph detection device |
| CN104382590A (en) * | 2014-12-11 | 2015-03-04 | 赖大坤 | Automatic shockable rhythm identification and classification method combined with electrocardio time-frequency domain feature analysis |
| CN104573458A (en) * | 2014-12-30 | 2015-04-29 | 深圳先进技术研究院 | Identity recognition method, device and system based on electrocardiogram signals |
Non-Patent Citations (1)
| Title |
|---|
| 王法强: "基于度量学习的人脸识别方法研究", 《中国优秀硕士学位论文全文数据库•信息科技辑》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017084546A1 (en) * | 2015-11-17 | 2017-05-26 | 安徽华米信息科技有限公司 | Wearable device-based user interest information determination method, device and wearable device |
| US10163528B2 (en) | 2015-11-17 | 2018-12-25 | Anhui Huami Information Technology Co., Ltd. | Determining user-interested information based on wearable device |
| GB2552035A (en) * | 2016-07-08 | 2018-01-10 | B-Secur Ltd | ECG Authentication method and apparatus |
| CN106485213A (en) * | 2016-09-27 | 2017-03-08 | 鲁东大学 | A kind of utilization electrocardiosignal carries out the feature extracting method of automatic identification |
| CN106485213B (en) * | 2016-09-27 | 2019-04-12 | 鲁东大学 | A kind of feature extracting method carrying out automatic identification using electrocardiosignal |
| CN108304746B (en) * | 2017-01-11 | 2022-09-13 | 北京三星通信技术研究有限公司 | Method and equipment for updating authentication reference information for electrocardio identity authentication |
| CN108304746A (en) * | 2017-01-11 | 2018-07-20 | 北京三星通信技术研究有限公司 | Method and apparatus of the update for the certification reference information of electrocardio authentication |
| CN106897693A (en) * | 2017-02-23 | 2017-06-27 | 深圳先进技术研究院 | A kind of personal identification method, apparatus and system |
| WO2018152724A1 (en) * | 2017-02-23 | 2018-08-30 | 深圳先进技术研究院 | Identity recognition method, device and system |
| CN107239684A (en) * | 2017-05-22 | 2017-10-10 | 吉林大学 | A kind of feature learning method and system for ECG identifications |
| CN107374614A (en) * | 2017-07-25 | 2017-11-24 | 海南聚能科技创新研究院有限公司 | The automatic classification storage control method of electrocardiogram and system |
| WO2021068107A1 (en) * | 2019-10-08 | 2021-04-15 | 北京微动数联科技有限公司 | Identity recognition method based on ballistocardiogram signal, electronic device, and storage medium |
| CN110869930A (en) * | 2019-10-08 | 2020-03-06 | 焦旭 | Identity recognition method based on cardiac shock signal, electronic device and storage medium |
| CN111063437A (en) * | 2019-12-12 | 2020-04-24 | 中科海微(北京)科技有限公司 | Personalized chronic disease analysis system |
| CN111063437B (en) * | 2019-12-12 | 2024-01-23 | 中科海微(北京)科技有限公司 | Personalized chronic disease analysis system |
| CN112257518A (en) * | 2020-09-30 | 2021-01-22 | 西安交通大学第二附属医院 | ECG identity recognition method based on WT and WOA-PNN algorithm |
| CN112401902A (en) * | 2020-12-04 | 2021-02-26 | 山东大学 | Electrocardio identity recognition method and system based on neural network time-frequency analysis combination |
| CN113221736A (en) * | 2021-05-11 | 2021-08-06 | 菏泽学院 | Electrocardio identity recognition method and system based on multi-feature collaborative non-negative matrix factorization |
| CN113440109A (en) * | 2021-08-02 | 2021-09-28 | 和也健康科技有限公司 | Human physiological parameter detection and monitoring system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105468951B (en) | 2019-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105468951A (en) | Method and device for identity recognition through electrocardiographic feature and wearable device | |
| CN105286853A (en) | Disease detection method and device based on wearable device and wearable device | |
| CN101856225B (en) | Method for detecting R wave crest of electrocardiosignal | |
| CN101828918B (en) | Electrocardiosignal R peak detection method based on waveform characteristic matching | |
| EP2459065B1 (en) | Method and apparatus for the analysis of a ballistocardiogram signal | |
| Labati et al. | ECG biometric recognition: Permanence analysis of QRS signals for 24 hours continuous authentication | |
| CN103284702A (en) | Electrocardiogram and pulse wave relation analysis method and method and device of fusion analysis | |
| CN107688553B (en) | Method for detecting electrocardiographic waveform characteristics based on wavelet transform and logistic regression algorithm | |
| Singh et al. | Correlation-based classification of heartbeats for individual identification | |
| CN111091116A (en) | Signal processing method and system for judging arrhythmia | |
| CN104063645A (en) | Identity recognition method based on ECG dynamic self-updating samples | |
| US10163528B2 (en) | Determining user-interested information based on wearable device | |
| CN110786849B (en) | Electrocardiosignal identity recognition method and system based on multi-view discriminant analysis | |
| CN112257518A (en) | ECG identity recognition method based on WT and WOA-PNN algorithm | |
| CN105997043A (en) | Pulse frequency extraction method based on wrist type wearable device | |
| CN110226919A (en) | Electrocardiosignal type detection method, apparatus, computer equipment and storage medium | |
| CN110123289A (en) | A kind of biometric discrimination method and relevant apparatus based on pulse wave | |
| Li et al. | A novel abnormal ECG beats detection method | |
| Akhbari et al. | Fiducial points extraction and characteristicwaves detection in ECG signal using a model-based bayesian framework | |
| Mukhopadhyay et al. | QRS complex identification using Hilbert transform, variable threshold and slope reversal approach | |
| Vuksanovic et al. | Analysis of human electrocardiogram for biometric recognition using analytic and ar modeling extracted parameters | |
| Nair et al. | Adaptive wavelet based identification and extraction of PQRST combination in randomly stretching ECG sequence | |
| CN106419884A (en) | Heart rate calculating method and system based on wavelet analysis | |
| CN114861706A (en) | Electrocardio identity recognition method based on quality evaluation and deep transfer learning | |
| Saini et al. | P-and T-wave delineation in ECG signals using support vector machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |