[go: up one dir, main page]

CN112472105B - A method and system for ECG identification based on kernel-based bounded discriminant analysis - Google Patents

A method and system for ECG identification based on kernel-based bounded discriminant analysis Download PDF

Info

Publication number
CN112472105B
CN112472105B CN202011521898.9A CN202011521898A CN112472105B CN 112472105 B CN112472105 B CN 112472105B CN 202011521898 A CN202011521898 A CN 202011521898A CN 112472105 B CN112472105 B CN 112472105B
Authority
CN
China
Prior art keywords
bounded
kernel
discriminant analysis
signal
dimensional
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
Application number
CN202011521898.9A
Other languages
Chinese (zh)
Other versions
CN112472105A (en
Inventor
杨公平
朱桂萍
孙启玉
李红超
张永忠
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Fengshi Information Technology Co ltd
Shandong University
Original Assignee
Shandong Fengshi Information Technology Co ltd
Shandong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shandong Fengshi Information Technology Co ltd, Shandong University filed Critical Shandong Fengshi Information Technology Co ltd
Priority to CN202011521898.9A priority Critical patent/CN112472105B/en
Publication of CN112472105A publication Critical patent/CN112472105A/en
Application granted granted Critical
Publication of CN112472105B publication Critical patent/CN112472105B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/117—Identification of persons
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G06F18/24133—Distances to prototypes
    • G06F18/24143—Distances to neighbourhood prototypes, e.g. restricted Coulomb energy networks [RCEN]

Landscapes

  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Medical Informatics (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • Surgery (AREA)
  • Molecular Biology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Psychiatry (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Physiology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

本发明属于心电信号身份识别领域，提供了一种基于核的有界判别分析的心电身份识别方法及系统。其中，基于核的有界判别分析的心电身份识别方法包括获取心电信号并提取单周期的心拍信号；通过核函数将心拍隐式地投影到高维空间，在高维空间中使用有界判别分析将心拍信号映射到低维子空间，提取单周期的心拍信号的低维特征；基于欧式距离从模板数据库中寻找最相似的模板，输出最相似的模板对应的用户为待识别用户；所述模板数据库中存储有单周期的心拍信号模板及其对应用户。

Figure 202011521898

The invention belongs to the field of electrocardiographic signal identification, and provides an electrocardiographic identification method and system based on kernel bounded discriminant analysis. Among them, the ECG identification method based on the kernel bounded discriminant analysis includes acquiring the ECG signal and extracting the single-cycle heartbeat signal; implicitly projecting the heartbeat into the high-dimensional space through the kernel function, and using bounded in the high-dimensional space The discriminant analysis maps the heartbeat signal to a low-dimensional subspace, and extracts the low-dimensional features of the single-cycle heartbeat signal. Based on the Euclidean distance, the most similar template is found from the template database, and the user corresponding to the most similar template is output as the user to be identified. The template database stores a single-cycle heart beat signal template and its corresponding users.

Figure 202011521898

Description

Electrocardio identity recognition method and system based on bounded discriminant analysis of kernel
Technical Field
The invention belongs to the field of electrocardiosignal identity recognition, and particularly relates to an electrocardio identity recognition method and system based on bounded discriminant analysis of a kernel.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The identification technology based on the electrocardiosignals is widely concerned. Electrocardiosignals are important physiological signals on human bodies and contain unique identity information of individuals. In addition, the electrocardiosignals generally can not change greatly along with the time, and along with the development of the micro sensor technology, the electrocardiosignals are more and more convenient to acquire. The electrocardiosignal has the characteristics of universality, uniqueness, stability, easiness in acquisition and the like, is a safer and more reliable identity recognition technology, and has a good application prospect. The inventor finds that the existing identification method is often based on a certain data distribution hypothesis, can not well extract the features with discrimination in the electrocardiosignals, has the problems of low discrimination of the electrocardio features, insufficient robustness and the like, and how to better extract the electrocardio identity identification features with discrimination is one of the problems to be solved urgently at present.
Disclosure of Invention
In order to solve at least one technical problem in the background art, the invention provides an electrocardiogram identity recognition method and system based on kernel bounded discriminant analysis, aiming at the nonlinear distribution of electrocardiogram signals, the kernel bounded discriminant analysis uses a polynomial kernel function, and the electrocardiogram signals are implicitly projected to a high-dimensional space with distinguishability, so that the distinguishing degree of the characteristics of the electrocardiogram signals is further improved, and the effect of electrocardiogram identity recognition is improved.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides an electrocardio identity recognition method based on bounded discriminant analysis of a kernel.
An electrocardio identity recognition method based on bounded discriminant analysis of a kernel comprises the following steps:
acquiring an electrocardiosignal and extracting a single-cycle heart beat signal;
the heart beat is implicitly projected to a high-dimensional space through a kernel function, a bounded discriminant analysis is used in the high-dimensional space to map the heart beat signal to a low-dimensional subspace, and the low-dimensional features of the single-cycle heart beat signal are extracted;
searching the most similar template from the template database based on the Euclidean distance, and outputting a user corresponding to the most similar template as a user to be identified; the template database stores a single-cycle heart beat signal template and a corresponding user thereof.
In a second aspect, the present invention provides a system for cardiac electrical identity recognition based on bounded discriminant analysis of kernels.
An electrocardiogram identity recognition system based on bounded discriminant analysis of a kernel, comprising:
the signal acquisition module is used for acquiring the electrocardiosignals and extracting the cardiac signals of a single cycle;
the characteristic extraction module is used for implicitly projecting the heart beat to a high-dimensional space through a kernel function, mapping the heart beat signal to a low-dimensional subspace in the high-dimensional space by using bounded discriminant analysis, and extracting the low-dimensional characteristic of the heart beat signal in a single period;
the identity recognition module is used for searching the most similar template from the template database based on the Euclidean distance and outputting the user corresponding to the most similar template as the user to be recognized; the template database stores a single-cycle heart beat signal template and a corresponding user thereof.
A third aspect of the invention provides a computer-readable storage medium.
A computer-readable storage medium, having stored thereon a computer program which, when executed by a processor, performs the steps in the method for cardiac electrical identity recognition based on bounded discriminatory analysis of kernels as described above.
A fourth aspect of the invention provides a computer apparatus.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method for cardiac electrical identity recognition based on bounded discriminant analysis by kernel as described above when executing the program.
Compared with the prior art, the invention has the beneficial effects that:
preprocessing a electrocardiosignal by filtering and denoising, and detecting the position of the vertex of an R wave by using a Pan-Tompkins algorithm so as to extract a heart beat; obtaining a set of subspaces' bases through a kernel-based bounded discriminant analysis using a training set heartbeat; for any heart beat, the kernel function and the subspace substrate are utilized to obtain the low-dimensional electrocardio characteristic vector with more distinctiveness.
The invention discloses a method for reducing dimension by embedding graphs in a traditional method, which is characterized in that each kind of data is assumed to accord with Gaussian distribution and not accord with the real condition of electrocardiosignals, when the electrocardiosignal data do not meet the assumption, the dispersion degree between classes can not be well described through the compactness degree in the classes.
Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flowchart of an electrocardiogram identity recognition method based on kernel-bounded discriminant analysis according to an embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example one
Referring to fig. 1, the method for identifying an electrocardiogram based on bounded discriminant analysis of kernels according to the present embodiment includes:
s101: acquiring an electrocardiosignal and extracting a single-cycle heart beat signal.
In a specific implementation, some noise interference exists when the original electrocardio signal is acquired, such as myoelectric interference, contact noise, baseline drift and the like.
The frequency of normal electrocardiosignals is generally distributed between 0.5 Hz and 100Hz, wherein the main energy is concentrated between 1Hz and 40Hz, and the electrocardiosignals have weaker amplitude and are sensitive to noise. According to the frequency range of the electrocardiosignals, the scheme adopts a fourth-order band-pass Butterworth filter to carry out denoising, and the low-frequency cut-off frequency and the high-frequency cut-off frequency of the filter are set to be 1Hz and 40Hz respectively.
S102: the heart beat is implicitly projected to a high-dimensional space through a kernel function, a bounded discriminant analysis is used in the high-dimensional space to map the heart beat signal to a low-dimensional subspace, and the low-dimensional features of the single-cycle heart beat signal are extracted.
The electrocardiosignals are non-stable cycle-like signals, and one electrocardiosignal record comprises a plurality of single-cycle heart beats. A complete heart beat sequentially comprises a P wave band, a QRS wave band and a T wave band, wherein the peak amplitude of the R wave is most prominent and can be used as a basis for extracting a single heart beat.
And detecting the position of the top point of the R wave in the electrocardiosignal by using a Pan-Tompkins algorithm. Firstly, the signal is down-sampled to 200Hz sampling frequency, noise, P wave and T wave are weakened by using a filter, and QRS wave complex in the electrocardiosignal is highlighted. And enhancing the QRS complex by a difference equation to obtain a difference signal, and carrying out a series of nonlinear operations such as point-by-point squaring and window sliding integration on the difference signal to obtain an integral signal. And determining that the QRS complex is detected when the amplitudes of the points on the differential signal and the integral signal simultaneously satisfy a threshold setting. And searching the maximum value of the signal amplitude within the QRS wave group range, namely the position of the top point of the R wave. Finally, a backtracking mechanism is used for preventing missed detection and false detection.
Taking the R wave vertex position detected by a Pan-Tompkins algorithm as a reference, and taking a certain number of points forwards and backwards according to the sampling frequency of the electrocardiosignals to form a heart beat. And calculating a plurality of heartbeats segmented from each electrocardiogram record, and screening out the heartbeats with too large distance from the average heart beat by an average value threshold method.
The heart beat is implicitly projected to a high-dimensional space through a kernel function based on the bounded discriminant analysis of the kernel, and the heart beat is mapped to a low-dimensional subspace in the high-dimensional space by using the bounded discriminant analysis, so that the distances between positive samples are smaller, and the distances between negative samples are larger. To achieve this goal, kernel-based bounded discriminant analysis uses a neighbor graph to describe the degree of compactness within a class and a penalty graph to describe the degree of dispersion between classes. Meanwhile, aiming at the problem that the electrocardiosignals are inseparable in a linear space, a polynomial kernel function is introduced based on bounded discriminant analysis of kernels, and the electrocardiosignals are implicitly projected to a high-dimensional space with distinguishability
Figure BDA0002849262520000051
N single-cycle heart beat signals with the length of m obtained by dividing electrocardiosignals are used as training data of a learning process of a bounded discriminant analysis method based on a kernel, wherein the heart beats from the same user are similar heart beats, and otherwise, the heart beats are different heart beats. Let X be (X)1，x2，....xN) For class C heartbeats from different subjects at position C, with each column xiIs the ith heartbeat in m-dimensional space. If xiAnd xjAll from the same subject, called (x)i，xj) Positive sample pairs, otherwise called negative sample pairs。
The bounded discriminant analysis method based on kernel uses polynomial kernel function to implicitly map the heart beat to a high-dimensional space with more separability
Figure BDA0002849262520000061
For any x in the input spaceiAnd xjThe polynomial kernel satisfies:
k(xi，xj)＝Φ(xi)TΦ(xj). (1)
where Φ is mapping the heart beat to space
Figure BDA0002849262520000063
Of the method in
Figure BDA0002849262520000064
The inner product calculation in the space is converted into the calculation in the original input space through the kernel function, so that the specific phi does not need to be calculated. The formula of the kernel function is
k(x，y)＝(xTy+1)2. (2)
The distance in the high-dimensional space is calculated using a kernel function, as follows
Figure BDA0002849262520000062
Finding x in a high dimensional space by the distance calculated by the kernel functioniK of (a)1And (3) establishing a neighbor graph G ═ { X, W } by using neighbor heartbeats from the same class, wherein W is a mark matrix of the neighbor graph, and searching XiK of (a)2Creating a penalty map G for each neighboring heartbeat from different classesp＝{X，WpIn which W ispIs a mark matrix of the penalty map.
By means of G and GpThe kernel-based bounded discriminant analysis method solves a low-dimensional subspace that can simultaneously maintain the neighborhood relationship within and between classes. Solving an objective function of the subspace base as
Figure BDA0002849262520000071
Wherein D is a diagonal matrix defined as
Figure BDA0002849262520000072
Must be satisfied when the objective function is minimized
K(Dp-Wp)Kα＝λK(D-W)Kα. (6)
And decomposing the generalized eigenvalues to obtain eigenvectors, and selecting the eigenvectors corresponding to the first d smallest eigenvalues to form a group of bases A of the low-dimensional subspace.
After the low-dimensional subspace substrate A obtained by using the bounded discriminant analysis based on the kernel, any electrocardiosignal x can be expressed as a low-dimensional subspace characteristic y, and the expression formula is
Figure BDA0002849262520000073
The dimensionality reduction process of the kernel-based bounded discriminant analysis is as follows:
inputting: training data matrix X of heartbeat, dimensionality d after dimensionality reduction, and heartbeat neighbor number k1，k2。
And (3) outputting: the low-dimensional feature y.
Step1 the inner product between the pairs of points is calculated and saved using equation (2).
Step2 calculating distance using equation (3) to find xiK of (a)1Intra-class neighbor and k2Inter-class neighbors.
Step3 establishing a neighbor graph G and a penalty graph GpCalculating diagonal matrices D and D using equation (5)p。
And Step4, performing generalized eigenvalue decomposition by using the formula (6) to calculate eigenvalues and eigenvectors.
And Step5, selecting the eigenvectors corresponding to the minimum d eigenvalues to form a group of substrates A in the subspace.
Step6, for any heart beat x, the low-dimensional feature of the heart beat is calculated by using the formula (7).
S103: searching the most similar template from the template database based on the Euclidean distance, and outputting a user corresponding to the most similar template as a user to be identified; the template database stores a single-cycle heart beat signal template and a corresponding user thereof.
In the specific implementation, when in recognition, for a newly input heart beat to be recognized, the kernel function and the subspace base are also used for obtaining the low-dimensional characteristics of the signal. And further searching the most similar template through Euclidean distance calculation, and outputting the user corresponding to the most similar template as the user to be identified.
According to the characteristics of electrocardiosignals, preprocessing the electrocardiosignals through filtering, positioning and extracting a monocycle heart beat according to the top point of an R wave, and further learning a low-dimensional subspace substrate by adopting a bounded discriminant analysis method based on a kernel. Any electrocardiosignal sample can be projected from a high-dimensional space to a low-dimensional space to obtain a low-dimensional electrocardiosignal characteristic with more distinctiveness, and finally, the similarity comparison is carried out on the extracted electrocardiosignal characteristic by using Euclidean distance measurement. A penalty graph is specially established based on bounded discriminant analysis of the kernel to describe the inter-class dispersion degree of the electrocardiosignals, so that the extracted low-dimensional electrocardio characteristics can simultaneously reserve the intra-class and inter-class neighborhood relationship of the electrocardiosignals. Meanwhile, aiming at the nonlinear distribution of the electrocardiosignals, a polynomial kernel function is used based on bounded discriminant analysis of kernels, the electrocardiosignals are implicitly projected to a high-dimensional space with more distinguishability, the distinguishing degree of the electrocardiosignal characteristics is further improved, and the effect of electrocardio identity recognition is improved.
Example two
The embodiment provides an electrocardiogram identity recognition system based on bounded discriminant analysis of a kernel, which comprises:
the signal acquisition module is used for acquiring the electrocardiosignals and extracting the cardiac signals of a single cycle;
the characteristic extraction module is used for implicitly projecting the heart beat to a high-dimensional space through a kernel function, mapping the heart beat signal to a low-dimensional subspace in the high-dimensional space by using bounded discriminant analysis, and extracting the low-dimensional characteristic of the heart beat signal in a single period;
the identity recognition module is used for searching the most similar template from the template database based on the Euclidean distance and outputting the user corresponding to the most similar template as the user to be recognized; the template database stores a single-cycle heart beat signal template and a corresponding user thereof.
Each module in the electrocardiogram identification system based on the bounded discriminant analysis of the kernel according to this embodiment corresponds to each step in the electrocardiogram identification method based on the bounded discriminant analysis of the kernel according to the first embodiment one to one, and the specific implementation process is the same, and will not be described here again.
EXAMPLE III
The present embodiment provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps in the method for cardiac electrical identity recognition based on bounded discriminative analysis of kernels as described above.
Example four
The embodiment provides a computer device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to realize the steps in the electrocardio identity recognition method based on the bounded discrimination analysis of the kernel as described above.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (9)

1.一种基于核的有界判别分析的心电身份识别方法，其特征在于，包括：获取心电信号并提取单周期的心拍信号；1. an electrocardiographic identification method based on nuclear bounded discriminant analysis, is characterized in that, comprises: obtain electrocardiographic signal and extract the heart beat signal of single cycle; 通过核函数将心拍隐式地投影到高维空间，在高维空间中使用有界判别分析将心拍信号映射到低维子空间，提取单周期的心拍信号的低维特征；所述提取单周期的心拍信号的低维特征的过程为：The heartbeat is implicitly projected to the high-dimensional space through the kernel function, and the heartbeat signal is mapped to the low-dimensional subspace using bounded discriminant analysis in the high-dimensional space, and the low-dimensional feature of the single-cycle heartbeat signal is extracted; The process of low-dimensional features of the heartbeat signal is: 通过核函数计算的距离，在高维空间中，寻找同一用户的任一心拍的设定数量来自同类的近邻心拍，建立近邻图；寻找同一用户的任一心拍的设定数量来自不同类的近邻心拍，建立惩罚图；Through the distance calculated by the kernel function, in the high-dimensional space, find the set number of any heart beat of the same user from the same kind of neighbor beats, and establish a neighbor graph; find the set number of any heart beat of the same user from different types of neighbors Heart beat, establish punishment map; 借助近邻图和惩罚图，基于核的有界判别分析方法求解一个能同时保持类内和类间邻域关系的低维子空间；With the help of nearest neighbor graph and penalty graph, a kernel-based bounded discriminant analysis method solves a low-dimensional subspace that can maintain both intra-class and inter-class neighborhood relationships; 基于欧式距离从模板数据库中寻找最相似的模板，输出最相似的模板对应的用户为待识别用户；所述模板数据库中存储有单周期的心拍信号模板及其对应用户。The most similar template is searched from the template database based on the Euclidean distance, and the user corresponding to the most similar template is output as the user to be identified; the template database stores a single-cycle heartbeat signal template and its corresponding user. 2.如权利要求1所述的基于核的有界判别分析的心电身份识别方法，其特征在于，获取心电信号之前，使用四阶带通巴特沃斯滤波器消除原始心电信号的噪声。2. the electrocardiographic identification method based on kernel-based bounded discriminant analysis as claimed in claim 1, is characterized in that, before obtaining electrocardiographic signal, use fourth-order band-pass Butterworth filter to eliminate the noise of original electrocardiographic signal . 3.如权利要求2所述的基于核的有界判别分析的心电身份识别方法，其特征在于，四阶带通巴特沃斯滤波器的截止频率为1-40Hz。3 . The method for ECG identification based on kernel-based bounded discriminant analysis as claimed in claim 2 , wherein the cut-off frequency of the fourth-order band-pass Butterworth filter is 1-40 Hz. 4 . 4.如权利要求1所述的基于核的有界判别分析的心电身份识别方法，其特征在于，基于R波顶点选取左右两侧指定长度的信号，得到一个单周期的心拍信号。4. The electrocardiographic identification method based on kernel-based bounded discriminant analysis as claimed in claim 1, characterized in that, based on the R-wave apex, signals of specified lengths on the left and right sides are selected to obtain a single-cycle cardiac beat signal. 5.如权利要求4所述的基于核的有界判别分析的心电身份识别方法，其特征在于，以Pan-Tompkins算法检测到的R波顶点位置为基准，根据心电信号的采样频率向前向后各取一定数目的点构成一个心拍。5. the electrocardiographic identification method based on the bounded discriminant analysis of kernel as claimed in claim 4, is characterized in that, with the R wave vertex position that Pan-Tumpkins algorithm detects as reference, according to the sampling frequency of electrocardiogram A certain number of points are taken from the front to the back to form a heart beat. 6.如权利要求1所述的基于核的有界判别分析的心电身份识别方法，其特征在于，还包括：对每条心电信号分割出的多个心拍，计算出平均心拍，通过平均值阈值法筛除与平均心拍距离过大的心拍。6. the electrocardiographic identification method based on the kernel-based bounded discriminant analysis as claimed in claim 1, is characterized in that, also comprises: to the multiple heart beats that each electrocardiogram signal is segmented out, calculate the average heart beat, by averaging The value threshold method filters out the beats that are too far away from the average beat. 7.一种基于核的有界判别分析的心电身份识别系统，其特征在于，包括：7. A kind of electrocardiographic identification system based on the bounded discriminant analysis of kernel, is characterized in that, comprises: 信号获取模块，其用于获取心电信号并提取单周期的心拍信号；a signal acquisition module, which is used for acquiring ECG signals and extracting single-cycle cardiac beat signals; 特征提取模块，其通过核函数将心拍隐式地投影到高维空间，在高维空间中使用有界判别分析将心拍信号映射到低维子空间，提取单周期的心拍信号的低维特征；所述提取单周期的心拍信号的低维特征的过程为：A feature extraction module, which implicitly projects the heartbeat to a high-dimensional space through a kernel function, uses bounded discriminant analysis in the high-dimensional space to map the heartbeat signal to a low-dimensional subspace, and extracts the low-dimensional features of the single-cycle heartbeat signal; The process of extracting the low-dimensional features of the single-cycle cardiac beat signal is as follows: 通过核函数计算的距离，在高维空间中，寻找同一用户的任一心拍的设定数量来自同类的近邻心拍，建立近邻图；寻找同一用户的任一心拍的设定数量来自不同类的近邻心拍，建立一个惩罚图；Through the distance calculated by the kernel function, in the high-dimensional space, find the set number of any heart beat of the same user from the same kind of neighbor beats, and establish a neighbor graph; find the set number of any heart beat of the same user from different types of neighbors Heart beat, build a punishment map; 借助近邻图和惩罚图，基于核的有界判别分析方法求解一个能同时保持类内和类间邻域关系的低维子空间；With the help of nearest neighbor graph and penalty graph, a kernel-based bounded discriminant analysis method solves a low-dimensional subspace that can maintain both intra-class and inter-class neighborhood relationships; 身份识别模块，其基于欧式距离从模板数据库中寻找最相似的模板，输出最相似的模板对应的用户为待识别用户；所述模板数据库中存储有单周期的心拍信号模板及其对应用户。The identity recognition module searches for the most similar template from the template database based on Euclidean distance, and outputs the user corresponding to the most similar template as the user to be identified; the template database stores a single-cycle heartbeat signal template and its corresponding users. 8.一种计算机可读存储介质，其上存储有计算机程序，其特征在于，该程序被处理器执行时实现如权利要求1-6中任一项所述的基于核的有界判别分析的心电身份识别方法中的步骤。8. A computer-readable storage medium on which a computer program is stored, characterized in that, when the program is executed by a processor, a kernel-based bounded discriminant analysis as claimed in any one of claims 1-6 is implemented. Steps in the ECG identification method. 9.一种计算机设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，其特征在于，所述处理器执行所述程序时实现如权利要求1-6中任一项所述的基于核的有界判别分析的心电身份识别方法中的步骤。9. A computer device, comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the processor implements any of claims 1-6 when the processor executes the program. Steps in a described method for ECG identification based on kernel bounded discriminant analysis.
CN202011521898.9A 2020-12-21 2020-12-21 A method and system for ECG identification based on kernel-based bounded discriminant analysis Active CN112472105B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011521898.9A CN112472105B (en) 2020-12-21 2020-12-21 A method and system for ECG identification based on kernel-based bounded discriminant analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011521898.9A CN112472105B (en) 2020-12-21 2020-12-21 A method and system for ECG identification based on kernel-based bounded discriminant analysis

Publications (2)

Publication Number Publication Date
CN112472105A CN112472105A (en) 2021-03-12
CN112472105B true CN112472105B (en) 2021-08-17

Family

ID=74915200

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011521898.9A Active CN112472105B (en) 2020-12-21 2020-12-21 A method and system for ECG identification based on kernel-based bounded discriminant analysis

Country Status (1)

Country Link
CN (1) CN112472105B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN111582138A (en) * 2020-04-30 2020-08-25 山东大学 Electrocardio identity recognition method and system based on frequency domain cepstrum coefficient characteristics

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004012388A1 (en) * 2002-07-29 2004-02-05 C-Signature Ltd. Method and apparatus for electro-biometric identiy recognition

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN111582138A (en) * 2020-04-30 2020-08-25 山东大学 Electrocardio identity recognition method and system based on frequency domain cepstrum coefficient characteristics

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于形态特征和KPCA融合特征的ECG身份识别;陈晓丹等;《电子技术》;20150325(第03期);第5-8页 *
基于核理论的非线性鉴别分析方法的研究;杨旭等;《电脑知识与技术》;20161231(第04期);全文 *

Also Published As

Publication number Publication date
CN112472105A (en) 2021-03-12

Similar Documents

Publication Publication Date Title
Houssein et al. ECG signals classification: a review
Liu et al. Real-time multilead convolutional neural network for myocardial infarction detection
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
CN106473750A (en) Personal identification method based on photoplethysmographic optimal period waveform
Henzel et al. Atrial fibrillation episodes detection based on classification of heart rate derived features
Li et al. Identifying the mislabeled training samples of ECG signals using machine learning
Liu et al. A novel P-QRS-T wave localization method in ECG signals based on hybrid neural networks
Velic et al. Computer aided ECG analysis—State of the art and upcoming challenges
Yang et al. A novel approach for heart ventricular and atrial abnormalities detection via an ensemble classification algorithm based on ECG morphological features
Kaya Classification of PVC beat in ECG using basic temporal features
Xia et al. Real-time ECG delineation with randomly selected wavelet transform feature and random walk estimation
Chandrashekhar et al. Pulse ID: the case for robustness of ECG as a biometric identifier
CN112472105B (en) A method and system for ECG identification based on kernel-based bounded discriminant analysis
Benabdallah et al. Biometric individual authentication system using high performance ECG fiducial features
Tantawi et al. Electrocardiogram (ECG): a new burgeoning utility for biometric recognition
Hameed et al. A survey on various machine learning approaches for human electrocardiograms identification
Imah et al. Electrocardiogram for biometrics by using Adaptive Multilayer Generalized Learning Vector Quantization (AMGLVQ): integrating feature extraction and classification
CN113017620A (en) Electrocardio identity recognition method and system based on robust discriminant non-negative matrix decomposition
Nankani et al. R-peak detection from ECG signals using fractal based mathematical morphological operators
Ma et al. Out-of-distribution Representation and Graph Neural Network Fusion Learning for ECG Biometrics
Albuquerque et al. Classification of electrocardiography signals for user authentication based on ensembles with random undersampling
Jang et al. Robust human identification using ECG: Eigenpulse revisited
Fu et al. ECG delineation with randomly selected wavelet feature and random forest classifier
Vuksanovic et al. Analysis of human electrocardiogram for biometric recognition using analytic and AR modeling extracted parameters
Vysiya et al. Automatic detection of cardiac arrhythmias in ECG signal for IoT application

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