CN112016405A - Electrocardiogram identity authentication method based on wearable equipment - Google Patents
Electrocardiogram identity authentication method based on wearable equipment Download PDFInfo
- Publication number
- CN112016405A CN112016405A CN202010788412.1A CN202010788412A CN112016405A CN 112016405 A CN112016405 A CN 112016405A CN 202010788412 A CN202010788412 A CN 202010788412A CN 112016405 A CN112016405 A CN 112016405A
- Authority
- CN
- China
- Prior art keywords
- electrocardiogram
- feature
- reference features
- identity authentication
- noise reduction
- 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
Images
Classifications
-
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Computer Security & Cryptography (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Human Computer Interaction (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 discloses an electrocardiogram identity authentication method based on wearable equipment. Aiming at the problem of electrocardiogram noise reduction on wearable equipment, the invention designs mixed three-layer noise reduction to effectively remove noise signals, the mixed three-layer noise reduction combines the advantages of morphological filtering and frequency domain filtering, the implementation method is simple and convenient, and the signal quality of the electrocardiogram obtained after filtering can be obviously ensured. The user characteristic template constructed by the invention reserves the respective hard index and soft index characteristics of the reference characteristic and the non-reference characteristic of the electrocardiogram, and ensures the authentication accuracy of the system.
Description
Technical Field
The invention belongs to the field of biological authentication, and particularly relates to an electrocardiogram identity authentication method based on wearable equipment.
Background
In recent years, wearable devices (such as smart watches) are widely applied in the fields of mobile payment, mobile communication, health information monitoring, smart home interaction and the like. Currently, there are smart watches with an electrocardiogram measurement function, such as Apple Watch and health Watch of the Chinese rice technology.
In the past, research data in electrocardiogram research are acquired by an electrocardiograph with high anti-interference performance and high precision, and the electrocardiogram data have high signal-to-noise ratio and do not have high-energy low-frequency noise and high-frequency noise. Therefore, the prior electrocardiogram recognition and authentication research can remove the noise in the electrocardiogram by only one-time band-pass filtering. However, such frequency domain filtering methods are not suitable for removing noise in electrocardiograms acquired by wearable devices. The electrocardiogram signal noise ratio that wearable equipment gathered is lower, and the key low frequency region of electrocardiogram has been covered to the noise frequency channel of high energy, and frequency domain filtering can't extract the frequency channel that represents the electrocardiogram signal from the frequency channel of signal and noise aliasing effectively. On the other hand, in constructing the user feature template, the conventional research usually ignores the characteristics of the electrocardiogram features and directly constructs the feature template, and few researches are made for distinguishing and processing the two types of features, and the specific gravity of the features in the user feature template is not adjusted, so that the constructed feature template cannot sufficiently represent the electrocardiogram information of the user.
Disclosure of Invention
The invention aims to provide an electrocardiogram identity authentication method based on wearable equipment, aiming at the defects of the prior art.
The purpose of the invention is realized by the following technical scheme: an electrocardiogram identity authentication method based on wearable equipment comprises the following steps:
the method comprises the following steps: the wearable device with the electrocardiogram measurement function acquires an electrocardiogram of a user;
step two: preprocessing the electrocardiogram collected in the first step by applying mixed three-layer noise reduction;
step three: carrying out feature point detection on the electrocardiogram signal preprocessed in the step three, and positioning the position of a feature point;
step four: extracting reference features and non-reference features in the electrocardiogram based on the feature points positioned in the step three, and combining the reference features and the non-reference features to construct a feature template to form a user database;
step five: constructing a support vector machine based on the user database obtained in the step four;
step six: and obtaining an electrocardiogram feature template of the user to be authenticated according to the first step, the second step and the third step, inputting the class of support vector machines constructed in the fifth step, and authenticating the identity.
Further, in the first step, the wearable device includes a smart band, a smart watch, and a health band.
Further, in the second step, the three-layer noise reduction mixture is sequentially a first non-local mean noise reduction, a frequency domain filtering method and a second non-local mean noise reduction.
Further, in the first non-local mean denoising, the region length L of the target point and the field pointΔSetting the size of a QRS complex wave area, wherein the size N (p) of a search area is not less than half of the length of an electrocardiogram signal, and a smoothing parameter lambda is not more than half of the standard deviation of signal noise; in the second non-local mean noise reduction, LΔThe size of the QRS complex wave area is N (p), the size of the search area is not less than half of the length of the electrocardiogram signal, and the smoothing parameter lambda is 30% -50% of the standard deviation of signal noise.
Further, the frequency domain filtering method includes a wavelet denoising method and an elliptic filter.
Further, the third step is specifically: firstly, detecting the position of QRS complex wave in electrocardiogram signals by a band-pass filter, a derivative filter, a square sum integrator by adopting a Pan-Tompkins algorithm, roughly positioning the position of the QRS complex wave, and then accurately positioning the positions of characteristic points P, Q, R, S and a T point by a local adaptive threshold method.
Further, the fourth step includes the following substeps:
(4.1) calculating a time interval and an amplitude difference between the electrocardiogram feature points P, Q, R, S and the T point as reference features;
(4.2) intercepting electrocardiogram segments of one period from the electrocardiogram signals by taking the characteristic point R as the center, and calculating the mean value of all the intercepted electrocardiogram segments to obtain an average electrocardiogram segment;
(4.3) extracting non-reference features from the average electrocardiogram fragment obtained in the step (4.2), and extracting feature information of the electrocardiogram in a wavelet domain as the non-reference features by adopting wavelet decomposition;
(4.4) performing feature selection on the reference features obtained in the step (4.1) by adopting a Fisher algorithm to obtain electrocardiogram reference features with the most distinguishing degree, and performing dimensionality reduction on the non-reference features by adopting a multi-dimensional scaling method; debugging the identity authentication recognition effect by adjusting the reference feature number and the non-reference feature number in the feature template; the feature templates of all users constitute a user database.
Further, the step (4.3) is specifically: and decomposing the average electrocardiogram fragment by using the wavelet base db5, wherein the number of decomposition layers is 4, and the approximation coefficient of the 4 th layer and the detail coefficients of the 3 rd and 4 th layers are taken as non-reference characteristics.
Further, in the fifth step, the kernel function of the support vector machine of one type is set as a gaussian kernel function, and the anomaly ratio is 0.006.
The invention has the beneficial effects that: aiming at the problems of electrocardiogram noise reduction and the defects of characteristic template construction in the prior art, the invention provides a new noise reduction mode and a characteristic template construction method; the problem of making an uproar falls in the electrocardiogram on the wearable equipment is solved, the attitude of heart characteristic in order to improve the authentication rate of accuracy of protection electrocardiogram. The proportion of the reference characteristic to the non-reference characteristic is comprehensively adjusted to obtain the electrocardiogram characteristic template with strong representativeness, the implementation method is simple, the means is flexible, and the high accuracy and the low false recognition rate of the authentication system can be ensured.
Drawings
FIG. 1 is a schematic view of a use scenario of the present invention;
FIG. 2 is a flow chart of a wearable device based electrocardiogram identity authentication method;
FIG. 3 is a flow chart of hybrid three-layer noise reduction;
fig. 4 is a flow chart of feature extraction.
Detailed Description
In order to further explain the technical means adopted by the invention and the obtained effects, the invention is explained in detail below according to the attached drawings.
The wearable equipment such as the smart watch with the electrocardiogram measuring function can acquire the electrocardiogram of the user at any time for real-time authentication, and can also be used as auxiliary authentication of the intelligent equipment to realize the functions of opening an entrance guard, unlocking a mobile phone, payment authentication and the like, wherein a specific situation diagram is shown in fig. 1.
As shown in fig. 2, fig. 2 is a flowchart of an electrocardiogram identity authentication method based on a wearable device, including the following steps:
the method comprises the following steps: the user wears wearable equipment such as intelligent bracelet, intelligent wrist-watch and healthy wrist strap that have heart electrograph measurement function, for example the user can wear the existing intelligent wrist-watch that has heart electrograph measurement function on the market in wrist department, and wearable equipment gathers user's original heart electrograph data.
In this step, the user contacts the collecting electrode on the wearable device, the wearable device is internally provided with an electrocardiogram collecting chip, the electrocardiogram signal can be captured through weak potential change of the skin on the electrode, and the collected original electrocardiogram signal is uploaded to the terminal or the server.
Step two: and (3) denoising the original electrocardiogram signal by using the mixed three-layer denoising to obtain an effective electrocardiogram signal.
This step is one of the core steps of the present invention, as shown in fig. 3, fig. 3 is a flow chart of hybrid three-layer noise reduction, which is divided into the following sub-steps:
1) the acquired electrocardiogram signals of the user contain aliasing low-frequency and high-frequency noises with high energy, three layers of noise reduction are mixed, namely, the original electrocardiogram signals are subjected to noise reduction for three times, and the original electrocardiogram signals are subjected to noise reduction by non-local mean noise reduction for the first time; non-local mean noise reduction takes advantage of the non-correlation of the noise signal with the electrocardiogram signal, as well as the strong correlation of the electrocardiogram signal, to extract the effective electrocardiogram signal from the aliased signal.
The low-frequency noise and the high-frequency noise of the first non-local mean noise reduction filtering part can not excessively smooth signals, can also save complete detail characteristics, and still contain more noise. The parameters of the first non-local mean noise reduction are respectively set as: λ 0.5 × 0.08, LΔ10 and n (p) 3000. Lambda is a smoothing parameter, the degree of the smoothed control signal is, the larger the lambda is, the more gradual the change of the Gaussian function is, the higher the noise reduction level is, but the excessive noise is easily causedThe degree is smooth, more details can be kept when the lambda is smaller, but excessive noise data can be remained, and the smoothing effect is not obvious enough; the value of lambda is related to signal noise, and the smoothing parameter lambda is positively related to the noise standard deviation sigma, namely, lambda is k sigma; when the standard deviation of the signal noise is too large, a larger lambda is needed to smooth the noise, and if the standard deviation of the signal noise is too small, a smaller lambda is needed to smooth the noise. The first non-local mean noise reduction parameter should be set to be gentle to prevent excessive smoothing, so that the lambda parameter needs to be controlled not to exceed half of the signal noise standard deviation sigma, and the first noise reduction is ensured not to lose the detail features. Region length L of target point and field pointΔThe size of the similar region of the electrocardiogram, i.e. the size of the QRS complex region, is set. N (p) determines the size of the search area, and N (p) is set to be not less than half of the length of the electrocardiogram signal so as to search more similar morphological areas.
2) And the second time of noise reduction adopts a frequency domain filtering method.
The high-frequency noise of the electrocardiogram can be filtered by a frequency domain filtering method, and the key low-frequency information of the electrocardiogram is protected; firstly, detail information of the electrocardiogram is filtered by a wavelet denoising method, and then high-frequency noise is further filtered by an elliptic low-pass filter; the frequency domain filtered electrocardiogram signal still contains low-frequency noise, but the detail features are completely preserved.
3) And the third noise reduction is carried out again by adopting non-local mean noise reduction to obtain a clean and effective electrocardiogram signal.
And denoising and removing low-frequency noise for the second non-local mean value, completely storing the detail characteristics of the electrocardiogram, and setting the parameters of denoising for the second non-local mean value as: λ 0.4 × 0.03, LΔ10 and n (p) 3000, LΔStill set to the size of the similar region of the electrocardiogram, i.e., the size of the QRS complex region, N (p) is still set to not less than half the length of the electrocardiogram signal, so as to search for more similar morphology regions. Because the electrocardiogram contains less noise after two times of denoising, the lambda parameter does not need to be too large, and the lambda parameter is 30% -50% of the standard deviation sigma of the signal noise.
This can remove aliasing electrocardiogram noise and effectively preserve the morphological features of the electrocardiogram. The electrocardiogram signals after noise reduction for three times are well filtered, and the details of the electrocardiogram are completely stored.
Step three: and positioning each characteristic point in the electrocardiogram after noise reduction.
In this step, a Pan-Tompkins algorithm is used in combination with a local adaptive thresholding method to locate ECG feature points. Firstly, the Pan-Tompkins algorithm detects the position of a QRS complex wave through a band-pass filter, a derivative filter, a square sum integrator, roughly positions the position of the QRS complex wave, and then accurately positions the position of a characteristic point through a local self-adaptive threshold method.
Step four: the method comprises the following steps of extracting reference features and non-reference features of the electrocardiogram based on feature points positioned in the third step, and constructing a feature template corresponding to the input electrocardiogram by combining the reference features and the non-reference features; as shown in fig. 4, fig. 4 is a flow chart of feature extraction, and the feature template construction method is divided into the following sub-steps:
1) extracting reference features by taking the feature points of the electrocardiogram as reference points, and extracting non-reference features based on average electrocardiogram segments;
after the feature point detection of the previous step, the feature points in each electrocardiogram segment are located, and the time interval and amplitude difference between the electrocardiogram feature points P, Q, R, S and the T point are calculated as the reference features. In the invention, 9 time interval features and 10 amplitude features (RQ amplitude features are removed) are extracted in total, and the details are shown in Table 1.
Table 1: 9 time interval features and 10 amplitude features of electrocardiogram
Before extracting non-reference features, an average electrocardiogram fragment must be calculated. From a series of electrocardiogram signals, according to the position of the characteristic point R, positioning the electrocardiogram positions, then taking the characteristic point R as the center to intercept a period of electrocardiogram segments, and calculating the average value of all the intercepted electrocardiograms to obtain an average electrocardiogram segment.
Extracting non-reference features on the average electrocardiogram segment, and extracting feature information of the electrocardiogram in a wavelet domain by adopting wavelet decomposition, wherein the method specifically comprises the following steps: the average electrocardiogram fragment is decomposed by wavelet base db5, the number of decomposition layers is 4, and the 4 th layer approximation coefficient and the 3 rd and 4 th layer detail coefficients are used as non-reference features.
2) And selecting the features of the reference features, reducing the dimension of the features of the non-reference features, and adjusting the number of the reference features and the number of the non-reference features to construct a feature template.
Selecting the features of the reference features by using a Fisher algorithm to select the features with the most distinguishing degree; and reducing the dimension of the high-dimensional non-reference features by adopting a multi-dimensional scaling method to extract core representative non-reference features, reducing feature redundancy and preventing high-dimensional disasters. According to electrocardiogram data provided by a user, a user feature template containing different reference features and non-reference feature combination matching is constructed, the reference feature number and the non-reference feature number in the feature template are adjusted to enable the user feature template to obtain comprehensive optimal performance when the user recall rate and the user misrecognition rate are tested, the most representative user feature template is obtained, and a user database is established. Respectively selecting and reducing the dimension of the reference characteristic and the non-reference characteristic, keeping the characteristics of respective soft and hard indexes, and constructing a complete user characteristic template; the user feature template constructed based on the reference features and the non-reference features not only retains the detail features of the electrocardiogram, but also stores the overall morphological features of the electrocardiogram.
Step five: and constructing a type of support vector machine for identity authentication according to the user database obtained in the step four.
In the step, a hyperplane is constructed by adopting a support vector machine of one type based on a user database to distinguish a user from a non-user, if the input features are within the hyperplane, the user is indicated, otherwise, the non-user is indicated; the kernel function of the support vector machine is set to be a gaussian kernel function, and the anomaly ratio is 0.006.
Step six: when identity authentication is carried out, obtaining an electrocardiogram feature template of a user to be subjected to identity authentication according to the first step, the second step and the third step, inputting the first support vector machine constructed in the fifth step, and carrying out identity authentication; the method is simple in calculation, fast in operation and suitable for being transplanted to each platform.
Claims (9)
1. An electrocardiogram identity authentication method based on wearable equipment is characterized by comprising the following steps:
the method comprises the following steps: a wearable device with an electrocardiogram measurement function acquires an electrocardiogram of a user.
Step two: and (3) preprocessing the electrocardiogram collected in the first step by applying mixed three-layer noise reduction.
Step three: and (4) carrying out feature point detection on the electrocardiogram signals preprocessed in the step three, so as to position the position of the feature point.
Step four: and extracting the reference features and the non-reference features in the electrocardiogram based on the feature points positioned in the step three, and combining the reference features and the non-reference features to construct a feature template to form a user database.
Step five: and constructing a type of support vector machine based on the user database obtained in the step four.
Step six: and obtaining an electrocardiogram feature template of the user to be authenticated according to the first step, the second step and the third step, inputting the class of support vector machines constructed in the fifth step, and authenticating the identity.
2. The wearable device-based electrocardiogram identity authentication method according to claim 1, wherein in the first step, the wearable device comprises a smart bracelet, a smart watch and a health bracelet.
3. The wearable device-based electrocardiogram identity authentication method according to claim 1, wherein in the second step, the three layers of noise reduction are mixed and sequentially comprise a first non-local mean noise reduction, a frequency domain filtering method and a second non-local mean noise reduction.
4. The wearable device-based electrocardiogram identity authentication method according to claim 3, wherein the first non-local mean noise reductionRegion length L of target point and field pointΔSetting the size of a QRS complex wave area, wherein the size N (p) of a search area is not less than half of the length of an electrocardiogram signal, and a smoothing parameter lambda is not more than half of the standard deviation of signal noise; in the second non-local mean noise reduction, LΔThe size of the QRS complex wave area is N (p), the size of the search area is not less than half of the length of the electrocardiogram signal, and the smoothing parameter lambda is 30% -50% of the standard deviation of signal noise.
5. The wearable device-based electrocardiogram identity authentication method according to claim 3, wherein said frequency domain filtering method comprises wavelet de-noising method and elliptic filter.
6. The wearable device-based electrocardiogram identity authentication method according to claim 1, wherein the third step is specifically: firstly, detecting the position of QRS complex wave in electrocardiogram signals by a band-pass filter, a derivative filter, a square sum integrator by adopting a Pan-Tompkins algorithm, roughly positioning the position of the QRS complex wave, and then accurately positioning the positions of characteristic points P, Q, R, S and a T point by a local adaptive threshold method.
7. The feature template construction method of claim 1, wherein the fourth step comprises the substeps of:
(4.1) calculating a time interval and an amplitude difference between the electrocardiogram feature points P, Q, R, S and the T point as reference features;
(4.2) intercepting electrocardiogram segments of one period from the electrocardiogram signals by taking the characteristic point R as the center, and calculating the mean value of all the intercepted electrocardiogram segments to obtain an average electrocardiogram segment;
(4.3) extracting non-reference features from the average electrocardiogram fragment obtained in the step (4.2), and extracting feature information of the electrocardiogram in a wavelet domain as the non-reference features by adopting wavelet decomposition;
(4.4) performing feature selection on the reference features obtained in the step (4.1) by adopting a Fisher algorithm to obtain electrocardiogram reference features with the most distinguishing degree, and performing dimensionality reduction on the non-reference features by adopting a multi-dimensional scaling method; debugging the identity authentication recognition effect by adjusting the reference feature number and the non-reference feature number in the feature template; the feature templates of all users constitute a user database.
8. The method of constructing a feature template according to claim 1, wherein the step (4.3) is embodied as: and decomposing the average electrocardiogram fragment by using the wavelet base db5, wherein the number of decomposition layers is 4, and the approximation coefficient of the 4 th layer and the detail coefficients of the 3 rd and 4 th layers are taken as non-reference characteristics.
9. The method according to claim 1, wherein in the fifth step, the kernel function of one type of support vector machine is set as a gaussian kernel function, and the anomaly ratio is 0.006.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010788412.1A CN112016405B (en) | 2020-08-07 | 2020-08-07 | Electrocardiogram identity authentication method based on wearable equipment |
| PCT/CN2020/122365 WO2022027830A1 (en) | 2020-08-07 | 2020-10-21 | Wearable device-based electrocardiogram identity authentication method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010788412.1A CN112016405B (en) | 2020-08-07 | 2020-08-07 | Electrocardiogram identity authentication method based on wearable equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112016405A true CN112016405A (en) | 2020-12-01 |
| CN112016405B CN112016405B (en) | 2022-07-15 |
Family
ID=73500174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010788412.1A Active CN112016405B (en) | 2020-08-07 | 2020-08-07 | Electrocardiogram identity authentication method based on wearable equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112016405B (en) |
| WO (1) | WO2022027830A1 (en) |
Citations (6)
| 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 |
| US20130261482A1 (en) * | 2012-03-28 | 2013-10-03 | Qualcomm Incorporated | Systems and methods for ecg monitoring |
| CN104102915A (en) * | 2014-07-01 | 2014-10-15 | 清华大学深圳研究生院 | Multiple-template matching identity recognition method based on ECG (Electrocardiogram) under electrocardiogram abnormality state |
| 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 |
| CN104706350A (en) * | 2014-11-13 | 2015-06-17 | 华中科技大学 | Electrocardio-signal noise reduction method |
| US20190147277A1 (en) * | 2016-04-29 | 2019-05-16 | Inesc Tec-Insituto De Engenharia De Sistemas E Computadores, Tecnologia E Ciência | Biometric method and device for identifying a person through an electrocardiogram (ecg) waveform |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102940485B (en) * | 2012-10-10 | 2015-04-08 | 广州三瑞医疗器械有限公司 | Algorithm for automatically recognizing fetal heart rate baseline |
| CN105748066B (en) * | 2016-03-03 | 2018-03-16 | 深圳竹信科技有限公司 | A kind of extracting method and device of electro-cardiologic signal waveforms characteristic point |
| WO2018152711A1 (en) * | 2017-02-22 | 2018-08-30 | 清华大学深圳研究生院 | Electrocardiographic authentication-based door control system and authentication method therefor |
-
2020
- 2020-08-07 CN CN202010788412.1A patent/CN112016405B/en active Active
- 2020-10-21 WO PCT/CN2020/122365 patent/WO2022027830A1/en active Application Filing
Patent Citations (6)
| 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 |
| US20130261482A1 (en) * | 2012-03-28 | 2013-10-03 | Qualcomm Incorporated | Systems and methods for ecg monitoring |
| CN104102915A (en) * | 2014-07-01 | 2014-10-15 | 清华大学深圳研究生院 | Multiple-template matching identity recognition method based on ECG (Electrocardiogram) under electrocardiogram abnormality state |
| CN104706350A (en) * | 2014-11-13 | 2015-06-17 | 华中科技大学 | Electrocardio-signal noise reduction method |
| 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 |
| US20190147277A1 (en) * | 2016-04-29 | 2019-05-16 | Inesc Tec-Insituto De Engenharia De Sistemas E Computadores, Tecnologia E Ciência | Biometric method and device for identifying a person through an electrocardiogram (ecg) waveform |
Non-Patent Citations (3)
| Title |
|---|
| HAO, H. ET.AL: "Multi-lead model-based ECG signal denoising by guided filter", 《JOURNAL OF MEDICAL AND BIOLOGICAL ENGINEERING》 * |
| 乔林峰等: "一种改进小波阈值的图像去噪算法", 《舰船电子工程》 * |
| 尹丽等: "采用集合经验模态分解和改进阈值函数的心电自适应去噪方法", 《西安交通大学学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112016405B (en) | 2022-07-15 |
| WO2022027830A1 (en) | 2022-02-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9870457B2 (en) | HERMA—heartbeat microwave authentication | |
| CN103345600B (en) | A kind of ecg signal data processing method | |
| Chawla | A comparative analysis of principal component and independent component techniques for electrocardiograms | |
| Li | Wavelets for electrocardiogram: overview and taxonomy | |
| WO2019192235A1 (en) | User identity authentication method and system based on mobile device | |
| CN105942974A (en) | Sleep analysis method and system based on low frequency electroencephalogram | |
| CN103610461A (en) | EEG noise elimination method based on dual-density wavelet neighborhood related threshold processing | |
| CN107293301B (en) | Recognition methods and system based on dental articulation sound | |
| CN113143261B (en) | Myoelectric signal-based identity recognition system, method and equipment | |
| CN117421539B (en) | Brain-computer interface system data optimization processing method | |
| CN115470832B (en) | Electrocardiosignal data processing method based on block chain | |
| Bastos et al. | Double authentication model based on ppg and ecg signals | |
| Yang et al. | Automated intra-patient and inter-patient coronary artery disease and congestive heart failure detection using EFAP-Net | |
| Chen et al. | Finger ECG based two-phase authentication using 1D convolutional neural networks | |
| CN108717535B (en) | Anesthesia depth estimation method based on mixed features and long-and-short-term memory network | |
| CN113468988B (en) | Identification method under multiple pressure states based on ECG (electronic pulse-taking) signals | |
| CN112016405B (en) | Electrocardiogram identity authentication method based on wearable equipment | |
| CN106691475B (en) | Emotion recognition model generation method and device | |
| CN109919050B (en) | Identity recognition method and device | |
| CN108197535A (en) | Refer to vein image quality evaluation method and device | |
| Bhavsar | Improved ECG Denoising using CEEMAN based on Complexity Measure and Nonlocal Mean Approach. | |
| CN111444489B (en) | Double-factor authentication method based on photoplethysmography sensor | |
| CN115989997A (en) | Sleep electroencephalogram signal staging method, equipment and medium for unsupervised clustering | |
| CN115299910A (en) | Privacy-protecting natural heart rate identification method | |
| CN113159205A (en) | Sparse time-frequency block common space mode feature extraction method based on optimal channel |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |