CN103345600B - A kind of ecg signal data processing method - Google Patents
A kind of ecg signal data processing method Download PDFInfo
- Publication number
- CN103345600B CN103345600B CN201310253575.XA CN201310253575A CN103345600B CN 103345600 B CN103345600 B CN 103345600B CN 201310253575 A CN201310253575 A CN 201310253575A CN 103345600 B CN103345600 B CN 103345600B
- Authority
- CN
- China
- Prior art keywords
- electrocardiosignal
- monocycle
- ecg signal
- signal data
- processing method
- 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.)
- Active
Links
Abstract
The invention discloses a kind of ecg signal data processing method, the method is comprised the following steps：Collection ecg signal data；Electrocardiogram (ECG) data to collecting is pre-processed, wherein the signal to gathering is removed baseline drift and denoising；Electrocardiosignal is decomposed into monocycle electrocardiosignal group；Feature extraction, according to the monocycle electrocardiosignal after normalization, Modulation recognition is carried out according to the similarity degree between curve signal, chooses maximum classification center telecommunications number as template；Detection signal is input into, and carries out similitude with centre data library template and compares, and confirms identity.List lead ecg signal data processing method of the invention, using indicatrix matching way, the similarity system design research of curve is carried out to electrocardiosignal sequence in itself, it is to avoid the complicated procedures of feature point extraction.
Description
【Technical field】
Data processing is carried out using electrocardiosignal the present invention relates to a kind of technology of identification, more particularly to one kind, is entered
And the method for recognizing identity.
【Background technology】
With the development of computer network and electronic technology, occur in that a kind of new auth method replaces traditional mouth
Order and password --- biometric identity identification technology.Biometric identity identification technology refers to be entered using human body biological characteristics or behavioural characteristic
A kind of technology of row authentication.
Automatic recognition system based on biological characteristic has the operation principle and the course of work being substantially the same.First it is collection
Sample, these samples can be image of fingerprint, face etc.；Next to that feature extraction is carried out, according to the uniqueness that sample has
With unique feature, with a kind of algorithm for it distributes a feature code, this feature code is stored into database.Finally work as needs
When carrying out identity authentication to someone, then be stored into certain Feature Correspondence Algorithm database this person feature code with it is identified
The feature of people matches, so as to find out its identity.
The method that electrocardiosignal of the prior art carries out identification, electrocardiosignal meets the basic bar of bio-identification
Part, the PQRST waveforms of the electrocardiogram of normal person keep relative constant within the regular hour, even if pressure, exercise heart rate
Change, but QRS wave shape remains in that stabilization, this ensures that theres the stability of individual ecg characteristics.But it is existing to be based on
Electrocardiosignal identity recognizing technology using the electrocardiosignal of 12 leads each lead multiple characteristic points, as with uniqueness and
The identification characteristic information of characteristic, the technology of this identification needs to gather the electrocardiosignal of 12 leads, signal acquisition
It is comparatively laborious, it is only applicable to implement by professional training hospital doctor, it is very difficult to apply in family and personal medical monitoring.Together
When, because the characteristic information for gathering is discrete data, calculate complicated and differentiation accuracy is not high.
【The content of the invention】
In order to solve the above technical problems, the present invention proposes a kind of ecg signal data processing method, and to the electrocardiosignal
Sequence carries out the similarity system design of indicatrix in itself：
A kind of ecg signal data processing method, the method is comprised the following steps：
A () gathers ecg signal data；
B () pre-processes to the ecg signal data for collecting；
C () feature extraction, will be decomposed into monocycle electrocardiosignal group by the electrocardiosignal of pretreatment, and by list therein
Cycle electrocardiosignal is used as a characteristic vector；
D () carries out categorizing selection to electrocardiosignal, identification characteristic vector is built with this；
E () similarity system design, confirms identity.
It is described to concretely comprise the following steps in features described above extraction step：
Detection electrocardiosignal QRS complex, determines R ripple wave crest points；
With crest as line of demarcation, electrocardiosignal is decomposed into monocycle signal group；
Using each monocycle signal as a characteristic vector.
Preferably, take and be in transverse axis (time shaft) and the longitudinal axis (voltage axis) enterprising line range simultaneously to each characteristic vector
[0,1] normalized.
Preferably, above-mentioned steps are additionally included in and cubic spline interpolation algorithm are carried out on transverse axis, and interpolation is at intervals of X=[0:
0.01:1]。
Preferably, in characteristic extraction step, further include by the monocycle electrocardiosignal after normalized
Classified, and selected the monocycle electrocardiosignal in maximum classification as set of eigenvectors.
Preferably, comprised the following steps before the similarity system design：
The monocycle electrocardiosignal extracted in maximum classification is stored in centre data as the set of eigenvectors of sample objects
Storehouse is used as original electro-cardiologic signals, storage to ATL；
Choose in monocycle electrocardiosignal classification, k normalization cardiac electrical cycle of similitude highest is used as personal representative
Signal, storage to ATL.
After detection signal input, similarity system design is carried out with centre data library template, determine identity.
Preferably, using indicatrix matching way, the similitude of curve is carried out in itself to monocycle electrocardiosignal sequence
Compare.
Advantage of the invention is that：
1st, ecg signal acquiring is carried out using family portable equipment, expands user scope, and be no longer limited to existing
Some ecg signal acquirings can only be completed by medical institutions such as hospitals；
2nd, by the way of the present invention is using indicatrix matching, it is to avoid the complicated processes of ecg signal data extraction so that
Calculate and simplify；
3rd, the signal data accuracy eliminated caused by the factors such as motion, mood, heart disease is not high, can effectively carry
The accuracy of identification high.
【Brief description of the drawings】
Fig. 1 is one embodiment center telecommunications number acquisition system structural representation；
Fig. 2 is the flow chart of one embodiment center telecommunications number collection；
Fig. 3 is original ecg signal data figure in one embodiment；
Fig. 4 is normalization electrocardio monocycle signal datagram in one embodiment；
Fig. 5 is the monocycle ecg signal data figure after normalization integration in one embodiment.
【Specific embodiment】
Below in conjunction with Figure of description 1-5, technical scheme of the present invention is elaborated further.
Fig. 1 and Fig. 2 respectively illustrate the present embodiment center telecommunications number acquisition system structural representation and flow chart, tool
Body is comprised the following steps：
Step S10, gathers electrocardiosignal.In the present embodiment, identification is carried out using single lead electrocardiosignal, wherein
Portable cardiac collecting device including medical 12 lead or other different lead numbers.
Due to it is existing based on electrocardiosignal identity recognizing technology using medical 12 lead electrocardiosignal each lead
Multiple characteristic points, signal acquisition is comparatively laborious, and range of application is smaller, therefore only need to use the heart of one of lead in the present embodiment
Electric data, can gather the data of different objects, different time sections, electrocardiosignal identification is no longer limited to hospital etc. and cure
Treat mechanism, using family portable electrocardiogram acquisition equipment, tele-medicine and individuality medical treatment, be voluntarily acquired by user and
Data are uploaded, range of application is expanded.
Step S20, the electrocardiogram (ECG) data to gathering is pre-processed.
ECG signal processing is mainly and is filtered, in the present embodiment, can using the fertile husband of the Bart of 0.5-45Hz with
Bandpass filter carries out denoising, and removes baseline drift using Wavelet Transformation Algorithm.
Step S30, the normalized of ecg signal data.
It is illustrated in figure 3 original ecg signal data figure in the present embodiment, the described pair of original electro-cardiologic signals number for collecting
According to being normalized, concretely comprise the following steps：
Determine monocycle electrocardiosignal and in this, as a characteristic vector.In the present embodiment, Fig. 4 show this implementation
Electrocardio monocycle signal datagram is normalized in example, wherein it is determined that being the step of monocycle signal in electrocardiosignal：
QRS wave in detection electrocardiosignal, and determine the position of R ripple wave crest points, with R ripple wave crest points as line of demarcation, by the heart
Electric signal is decomposed into monocycle electrocardiosignal group, using each monocycle electrocardiosignal as a characteristic vector.
It is simultaneously the normalizing of [0,1] in transverse axis (time shaft) and the longitudinal axis (voltage axis) enterprising line range to each characteristic vector
Change is processed, and cubic spline interpolation treatment is carried out on transverse axis, and interpolation interval is set as X=[0:0.01:1], so so that
Unified sample frequency is used in same information processing system.
The electrocardio exemplary plot after normalization is integrated is illustrated in figure 5, wherein, red straight line is interval for the segmentation of R crests, green
Color dot is T ripple crests, and red point is P ripple crests.With R ripples as line of demarcation, electrocardiosignal remains complete cycle after normalization
All features of signal, and amplitude difference disappears, and the electrocardiosignal after restructuring still can be very good to extract P ripples or R ripples
Information.
Step S40, sets up electrocardiosignal masterplate database.
In the present embodiment, as shown in figure 4, upper partial data is training set data, bottom is divided into test set data, by two
The data that part collects are stored in ecg signal data ATL as original electro-cardiologic signals data, upper partial data.
As shown in figure 4, bottom is divided into normalization monocycle electrocardio recall signal.For the original electro-cardiologic signals number for retrieving
According to by the monocycle confidence electric signal obtained after normalized, signal being carried out according to the similarity degree between curve signal
Classification, classification treatment is carried out using the method for cluster analysis, is chosen maximum class center electric signal and is known as personal identity
Other electrocardiosignal masterplate.Ecg signal data sampling is carried out by different objects, electrocardiosignal template database is set up.
In step S40, it would however also be possible to employ the intermediate value of all electrocardiosignals is calculated as standard vector using in maximum classification
Difference degree (such as Euclidean distance) between standard vector, and nearest 10 groups vectors of selected distance standard vector are used as adopting
The representative electrocardiosignal template of sample object.
Step S50, similarity system design.
Whether the characteristic vector for comparing the electrocardiosignal for collecting is similar to the template in electrocardiosignal template database, if
It is, then into step S60；If it is not, then terminating.
In the present embodiment, after detection electrocardiosignal input, indicatrix matching way is taken, entered with centre data library template
Row similarity system design.
Step S60：The similarity system design, is adopted with corresponding with the template database data that retrieval data similarity degree is maximum
Sample object, as the owner of retrieval electrocardiogram (ECG) data.
It should be noted that the specific embodiment of invention described above, is not constituted to the scope of the present invention
Restriction.Any technology according to the present invention design done other various corresponding changes and deformation, should be included in this
Within invention scope of the claims.
Claims (5)
1. a kind of ecg signal data processing method, it is characterised in that comprise the following steps：
A () gathers ecg signal data；
B () pre-processes to the ecg signal data for collecting；
C () feature extraction, will be decomposed into monocycle electrocardiosignal group by the electrocardiosignal of pretreatment, and by the monocycle therein
Electrocardiosignal further includes to be classified through the monocycle electrocardiosignal after normalized as a characteristic vector,
Classification treatment is carried out using the method for cluster analysis, and selects the monocycle electrocardiosignal in maximum classification as characteristic vector
Collection；The normalized is specially and is horizontally and vertically going up to enter each characteristic vector at the normalization that line range is [0,1]
Reason, wherein transverse axis are time shaft, and the longitudinal axis is voltage axis；
D () carries out categorizing selection to electrocardiosignal, identification characteristic vector is built with this；
E () similarity system design, confirms identity.
2. ecg signal data processing method as claimed in claim 1, it is characterised in that the characteristic extraction step is specific
For：
Detection electrocardiosignal QRS complex, determines R ripple wave crest points；
With crest as line of demarcation, electrocardiosignal is decomposed into monocycle signal group；
Using each monocycle signal as a characteristic vector.
3. ecg signal data processing method as claimed in claim 2, it is characterised in that be additionally included in is carried out three times on transverse axis
Spline interpolation treatment, interpolation is at intervals of X=[0:0.01:1].
4. ecg signal data processing method as claimed in claim 3, it is characterised in that include before the similarity system design
Following steps：
The monocycle electrocardiosignal extracted in maximum classification is made as the set of eigenvectors storage of sample objects in central database
It is original electro-cardiologic signals, storage to ATL；
Choose in monocycle electrocardiosignal classification, similitude highest k normalizes cardiac electrical cycle as personal representation signal,
Store ATL；
After detection signal input, similarity system design is carried out with centre data library template.
5. ecg signal data processing method as claimed in claim 4, it is characterised in that
Using indicatrix matching way, the similarity system design of curve is carried out in itself to monocycle electrocardiosignal sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310253575.XA CN103345600B (en) | 2013-06-24 | 2013-06-24 | A kind of ecg signal data processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310253575.XA CN103345600B (en) | 2013-06-24 | 2013-06-24 | A kind of ecg signal data processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103345600A CN103345600A (en) | 2013-10-09 |
| CN103345600B true CN103345600B (en) | 2017-06-13 |
Family
ID=49280395
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310253575.XA Active CN103345600B (en) | 2013-06-24 | 2013-06-24 | A kind of ecg signal data processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103345600B (en) |
Families Citing this family (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103610457B (en) * | 2013-11-19 | 2016-03-30 | 深圳先进技术研究院 | A kind of processing method of electrocardiosignal and system |
| CN103714281B (en) * | 2013-12-12 | 2017-03-29 | 深圳先进技术研究院 | A kind of personal identification method based on electrocardiosignal |
| CN103800003B (en) * | 2014-01-21 | 2016-10-05 | 杭州电子科技大学 | A kind of ECG detecting method and detector |
| CN103886204B (en) * | 2014-03-24 | 2017-10-17 | 宇龙计算机通信科技(深圳)有限公司 | Data storage device and date storage method |
| CN104102915B (en) * | 2014-07-01 | 2019-02-22 | 清华大学深圳研究生院 | Personal identification method based on ECG multi-template matching under a kind of anomalous ecg state |
| KR102257300B1 (en) * | 2014-09-01 | 2021-05-27 | 삼성전자주식회사 | Method and apparatus for authenticating user using ecg signal |
| CN104398252A (en) * | 2014-11-05 | 2015-03-11 | 深圳先进技术研究院 | Electrocardiogram signal processing method and device |
| CN104573458B (en) * | 2014-12-30 | 2017-05-31 | 深圳先进技术研究院 | A kind of personal identification method based on electrocardiosignal, apparatus and system |
| CN105469037A (en) * | 2015-11-19 | 2016-04-06 | 中科院微电子研究所昆山分所 | Identity identification method and identity identification system based on electrocardiosignals |
| CN106650685B (en) * | 2016-12-29 | 2020-06-05 | 深圳先进技术研究院 | Identity recognition method and device based on electrocardiogram signal |
| WO2018152724A1 (en) * | 2017-02-23 | 2018-08-30 | 深圳先进技术研究院 | Identity recognition method, device and system |
| CN107590508A (en) * | 2017-08-25 | 2018-01-16 | 漫迪医疗仪器（上海）有限公司 | A kind of mcg-signalses adaptive processing method and system |
| CN107736888A (en) * | 2017-11-27 | 2018-02-27 | 中山大学 | A kind of method and system that real-time cardiac monitoring is carried out using wearable ecg equipment |
| CN108875344A (en) * | 2018-06-25 | 2018-11-23 | Oppo（重庆）智能科技有限公司 | A kind of electronic equipment control method, electronic appliance controlling device and electronic equipment |
| CN109044347B (en) * | 2018-07-11 | 2021-08-27 | 上海夏先机电科技发展有限公司 | Method, device and system for identifying junctional escape of electrocardiowave image and electronic equipment |
| CN109492659B (en) * | 2018-09-25 | 2021-10-01 | 维灵（杭州）信息技术有限公司 | Method for calculating curve similarity for electrocardio and electroencephalogram waveform comparison |
| CN109222963A (en) * | 2018-11-21 | 2019-01-18 | 燕山大学 | A kind of anomalous ecg method for identifying and classifying based on convolutional neural networks |
| CN109893119B (en) * | 2019-03-24 | 2022-01-04 | 浙江好络维医疗技术有限公司 | P wave identification and positioning method based on multi-lead clustering analysis |
| CN110123289B (en) * | 2019-04-08 | 2022-01-07 | 清华大学深圳研究生院 | Biological identification method based on pulse wave and related device |
| CN110269625B (en) * | 2019-05-31 | 2022-02-11 | 杭州电子科技大学 | Novel multi-feature fusion electrocardio authentication method and system |
| CN110263520A (en) * | 2019-06-21 | 2019-09-20 | 深圳市奋达智能技术有限公司 | A kind of auth method and system |
| CN110169767B (en) * | 2019-07-08 | 2021-09-21 | 河北大学 | Retrieval method of electrocardiosignals |
| CN112370065A (en) * | 2020-10-21 | 2021-02-19 | 华南理工大学 | Identity recognition method, system and medium based on electrocardiosignal dynamics characteristics |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101773394B (en) * | 2010-01-06 | 2011-09-07 | 中国航天员科研训练中心 | Identification method and identification system using identification method |
| CN102274029B (en) * | 2011-05-25 | 2013-10-02 | 中国科学院深圳先进技术研究院 | Identity recognition method and system |
| CN102805620B (en) * | 2012-08-16 | 2014-03-05 | 深圳邦健生物医疗设备股份有限公司 | Method and device for detecting atrial fibrillation |
| CN102938055B (en) * | 2012-10-09 | 2015-12-02 | 哈尔滨工程大学 | A kind of hand bone identification system |
-
2013
- 2013-06-24 CN CN201310253575.XA patent/CN103345600B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN103345600A (en) | 2013-10-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103345600B (en) | A kind of ecg signal data processing method | |
| Labati et al. | Deep-ECG: Convolutional neural networks for ECG biometric recognition | |
| CN103714281B (en) | A kind of personal identification method based on electrocardiosignal | |
| Wieclaw et al. | Biometrie identification from raw ECG signal using deep learning techniques | |
| CN106214145B (en) | Electrocardiogram classification method based on deep learning algorithm | |
| CN101773394B (en) | Identification method and identification system using identification method | |
| Wang et al. | Arrhythmia classification algorithm based on multi-head self-attention mechanism | |
| CN109645980A (en) | A kind of rhythm abnormality classification method based on depth migration study | |
| CN109770862A (en) | Electrocardiosignal classification method, device, electronic equipment and storage medium | |
| Shen et al. | The PLR-DTW method for ECG based biometric identification | |
| Sepahvand et al. | A novel multi-lead ECG personal recognition based on signals functional and structural dependencies using time-frequency representation and evolutionary morphological CNN | |
| KR101578167B1 (en) | self-organized real-time authentication method using ECG signal | |
| Mingai et al. | A novel EEG feature extraction method based on OEMD and CSP algorithm | |
| El Boujnouni et al. | A wavelet-based capsule neural network for ECG biometric identification | |
| Hsu et al. | Exploring seismocardiogram biometrics with wavelet transform | |
| Kuila et al. | Feature extraction of electrocardiogram signal using machine learning classification | |
| Liu et al. | A novel P-QRS-T wave localization method in ECG signals based on hybrid neural networks | |
| CN109919050B (en) | Identity recognition method and device | |
| Wan et al. | Research on Identification Algorithm Based on ECG Signal and Improved Convolutional Neural Network | |
| CN114239649B (en) | Identity recognition method for discovering and recognizing new user by photoelectric volume pulse wave signal of wearable device | |
| CN206852594U (en) | A kind of device that user characteristics is obtained according to human-body biological electromagnetic wave | |
| CN114343679A (en) | Surface electromyogram signal upper limb action recognition method and system based on transfer learning | |
| Hsiao et al. | Multiple biometric authentication for personal identity using wearable device | |
| Tantawi et al. | On the use of the electrocardiogram for biometrie authentication | |
| CN113017620A (en) | Electrocardio identity recognition method and system based on robust discriminant non-negative matrix decomposition |
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 | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20131009 Assignee: Guangdong Xinshen Electronic Technology Co.,Ltd. Assignor: SHENZHEN INSTITUTES OF ADVANCED TECHNOLOGY CHINESE ACADEMY OF SCIENCES Contract record no.: X2023980039226 Denomination of invention: A Method for Processing ECG Signal Data Granted publication date: 20170613 License type: Common License Record date: 20230808 |
|
| EE01 | Entry into force of recordation of patent licensing contract |