CN111222498A - Identity recognition method based on photoplethysmography - Google Patents
Identity recognition method based on photoplethysmography Download PDFInfo
- Publication number
- CN111222498A CN111222498A CN202010194774.8A CN202010194774A CN111222498A CN 111222498 A CN111222498 A CN 111222498A CN 202010194774 A CN202010194774 A CN 202010194774A CN 111222498 A CN111222498 A CN 111222498A
- Authority
- CN
- China
- Prior art keywords
- size
- layer
- lstm
- photoplethysmography
- data
- 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.)
- Pending
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
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an identity recognition method based on photoelectric volume pulse waves, which comprises the following steps of 1) obtaining training group data and testing group data; 2) dividing all photoelectric volume pulse wave signal data in training group data and test group data into a plurality of segments containing a plurality of photoelectric volume pulse waves; 3) converting all the segmented segments into a time-frequency characteristic energy graph form by using continuous wavelet transform; 4) building a neural network model combining CNN and LSTM; 5) sending the picture data of the training group into a neural network model combining CNN and LSTM for training; 6) and classifying the test group picture data by using the trained model, wherein the classification result is an identity recognition result, and finally evaluating the trained neural network model combining the CNN and the LSTM. The method has high safety factor in application, does not need to manually extract features, has high and stable identification effect, and can be used for personal identification in the fields of government institutions, financial institutions and the like.
Description
Technical Field
The invention relates to the crossing field of information processing and computer science, in particular to an identity identification method based on photoelectric volume pulse waves, which can be used as a means for maintaining personal information security in the fields of government institutions, financial institutions and the like.
Background
With the development of the current society, the security problem becomes more prominent, people lose the reliability and the practicability of the traditional identity identification method by memorizing complex passwords or carrying electronic ciphers, and the current situation makes the demand of people on the biological identification technology more and more great. Such as company confidential systems, financial transactions, computer networks and access systems to the security domain are still identified and authorized by means of identification cards or passwords. Such systems are not sufficiently secure because identification card or password information is easily stolen or forgotten. The biological recognition system can identify according to the physiological signals and the behavior characteristics of the individual, and the physiological signals and the behavior characteristics of the human body are unique to the individual, so that the biological recognition system can identify and provide more secrecy and safety. Methods in which human physiological signals and behavioral features such as fingerprints, human faces, sounds, electroencephalograms, and electrocardiograms are used for identification are becoming increasingly popular. However, fingerprints can be copied by various means such as a powder method and a magnetic powder method, face recognition can be deceived by fake moving pictures, sounds can be simulated, and methods based on brain waves or electrocardiosignals require professional acquisition equipment and thus cannot be widely used.
Photoplethysmography (PPG) enables photoplethysmography to be used to obtain photoplethysmography from fingertip, wrist or earlobe measurements. PPG is a non-invasive, electro-optical method of obtaining information about the volume change of blood flow in a blood vessel by testing a part of the body close to the skin. The photoplethysmography signal is an inherent physiological signal of a human body, is difficult to copy and simulate, has high safety and is simple and convenient to acquire. Most of the existing identity recognition methods based on photoplethysmography need manual feature extraction, the process is complicated, features are greatly different due to different human bodies, and the generalization capability is low.
The time-frequency analysis technology can enable a user to observe the energy density and the intensity of a time domain and a frequency domain of a signal at the same time, and the time and the frequency are combined together, so that the signal can be fully processed. Continuous Wavelet Transform (CWT) is a time-frequency analysis technique, and is suitable for processing a non-stationary weak physiological signal, such as a photoplethysmographic pulse wave, mainly containing low-frequency components, and the information in the original signal can be better retained by using a time-frequency characteristic energy map obtained by the method. In addition, compared with the one-dimensional signal, the two-dimensional image can automatically ignore small noise data in the image by utilizing the convolution layer and the pooling layer in the model, thereby avoiding the influence of noise in the one-dimensional signal on the model identification accuracy and sensitivity.
The Convolutional Neural Network (CNN) is an algorithm which is good at processing a large amount of picture information, and mainly comprises an input layer, a Convolutional layer, a pooling layer, a full-link layer and a Softmax layer, wherein the CNN can retain spatial information, and a convolution structure is formed by the Convolutional layer and the pooling layer, so that the problem of overfitting of a mathematical model is effectively relieved. The Long Short-Term Memory Network (LSTM) is a special type in a Recurrent Neural Network (RNN), wherein the state of each unit is interacted with other units, the time dynamics in data is displayed through the internal feedback state, Long-Term dependence information can be learned, and the classification effect can be greatly improved by combining the CNN and the LSTM. Most of the existing identity recognition methods based on photoplethysmography are based on features extracted manually, which causes the problem of the difference of photoplethysmography signals among different human individuals, and leads to the reduction of the recognition accuracy and generalization capability of the model. The neural network model combining the CNN and the LSTM does not need to manually extract features, and the hidden and non-simulative features in the deep layer of the photoplethysmographic signal can be learned in the process of continuously fitting and continuously optimizing model parameters based on a large amount of data, so that the obtained model has high safety factor and generalization capability in practical application.
Disclosure of Invention
The invention aims to provide an identity identification method based on photoplethysmography aiming at the defects in the prior art. The method does not need to manually extract features, greatly simplifies the process of model fitting, and has the advantages of strong generalization capability, high safety coefficient, high and stable identification effect precision.
The technical scheme for realizing the purpose of the invention is as follows:
an identity recognition method based on photoplethysmography comprises the following steps:
1) obtaining training group data and test group data: collecting photoplethysmographic signals of n persons in a specified time period to form a training group; randomly acquiring photoelectric volume pulse wave signals of about 3/10 of the n persons in another time period to form a test group, wherein the photoelectric volume pulse wave signal data in the training group and the test group are identities of each person, and the data classification is identity recognition;
2) dividing all photoelectric volume pulse wave signal data in training group data and test group data into a plurality of segments containing a plurality of photoelectric volume pulse waves;
3) converting all the segmented segments into a time-frequency characteristic energy graph form by using continuous wavelet transform;
4) building a neural network model combining CNN and LSTM;
5) sending the picture data of the training group in the step 3) into a neural network model combining CNN and LSTM for training;
6) classifying the test group picture data in the step 3) by using the neural network model combining the CNN and the LSTM trained in the step 5), wherein the classification result is an identity recognition result, and finally evaluating the neural network model combining the CNN and the LSTM trained.
The number of the test group in the step 1) can be 2/10 to 4/10 of the total number of the participants, and the photoplethysmography data of each person needs to be collected as much as possible.
The segment length range of the photoplethysmography in the step 2) is between 5s and 20s, wherein the number of sampling points in the segment of the photoplethysmography is equal to the product of the sampling frequency and the length of the segmented photoplethysmography wave segment, and the time interval of the sampling points is 1 divided by the sampling frequency.
Selecting any one of 'cgau 8', 'haar', 'dB 2', 'bior' and 'mor 1' as a mother wavelet function of the continuous wavelet transform in the step 3), wherein the pixel size of the converted picture is 1054x148, and the picture is an RGB color image with a channel of 3.
The process of building the neural network model combining the CNN and the LSTM in the step 4) is as follows:
4.1, using the time-frequency characteristic energy graph with the uniform format as input layer data;
4.2 build up the convolution layer and the pooling layer, and the specific parameters are as follows:
a first layer: the number of filters is 30, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a second layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
and a third layer: the number of filters is 60, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a fourth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
and a fifth layer: the number of filters is 90, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a sixth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
a seventh layer: the number of filters is 120, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
an eighth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
a ninth layer: the number of filters is 150, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a tenth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
wherein the activation function in each convolutional layer is a ReLU activation function;
4.3 build the full connection layer and the LSTM layer, the concrete parameters are: the eleventh layer and the twelfth layer are full connection layers, the number of the neurons is 500 and 100 respectively, and the activation functions are ReLU activation functions; the thirteenth layer is an LSTM layer, where output _ dim is 50 in size; the fourteenth layer is a full connection layer, the number of the neurons is n, n is the number of the training groups, and the activation function is a Softmax activation function.
And the evaluation in the step 6) is to comprehensively evaluate the trained neural network model combining the CNN and the LSTM by using the loss function value and the accuracy.
The technical scheme has the advantages that:
1. the technical scheme adopts a deep learning method of a neural network model combining CNN and LSTM, does not need to independently extract manual features, does not need to carry out preprocessing such as denoising on collected photoplethysmography signals, directly combines the steps of feature extraction and classification identification, and simplifies the process of model fitting;
3. in the technical scheme, the characteristics which are hidden in the deep layer of the photoplethysmogram signal and cannot be simulated can be learned in the process of continuously fitting and continuously optimizing model parameters based on a large amount of data, so that the obtained model has high safety factor and generalization capability in practical application;
4. according to the technical scheme, the one-dimensional photoplethysmographic signals are converted into the time-frequency characteristic energy diagram, small noise data in the diagram can be ignored, and finally the small noise data are sent into the neural network model combining the CNN and the LSTM to achieve high-precision and stable identification effect.
The method has high safety factor in practical application, does not need manual feature extraction, greatly simplifies the process of model fitting, and has strong generalization capability and high and stable recognition effect precision.
Drawings
FIG. 1 is a schematic flow chart of an embodiment;
FIG. 2 is a graph of the time-frequency characteristic of 10s photoplethysmography waves of the same person at different time periods in the embodiment;
FIG. 3 is a graph of the time-frequency characteristic energy of 10s photoplethysmograph pulse waves of different persons in the example;
FIG. 4 is a diagram showing a neural network model combining CNN and LSTM in the embodiment;
FIG. 5 is a diagram illustrating the result of the photoplethysmography signal test in the embodiment.
In the figure, epoch is the number of iterations, loss and accuracycacy are the accuracy and loss function value of the simulation test, train _ loss is the loss function value of the training data, val _ loss is the loss function value of the test data, train _ accuracycacy is the accuracy of the training data, and val _ accuracycacy is the accuracy of the test data.
Detailed Description
The invention is described in further detail below with reference to the following figures and specific examples, but the invention is not limited thereto.
Example (b):
referring to fig. 1, an identity recognition method based on photoplethysmography includes the following steps:
1) downloading photoplethysmography signal data of 80 subjects from a MIMIC-III public database so as to simulate the process of acquiring the photoplethysmography signal data from a human body, and forming the downloaded photoplethysmography signals of 80 persons in the first 60 minutes into a training set; randomly extracting the photoplethysmography signals of 32 of the 80 persons in the last 60 minutes to form a test group, wherein the photoplethysmography signal data in the training group and the test group are the identities of the persons, and the data classification is identity recognition;
2) dividing all photoplethysmography signal data into a plurality of segments containing a plurality of photoplethysmography waves, wherein the length range of each segment is between 5s and 20s, the number of sampling points in the photoplethysmography segment is equal to the product of the sampling frequency and the length of the divided photoplethysmography segment, in the example, the sampling frequency is 125Hz, the length of the divided segment is 10s, and therefore the number of sampling points of one signal segment is 1250;
3) converting all the segmented segments into a time-frequency characteristic energy graph form by using continuous wavelet transform, wherein the continuous wavelet transform selects a mother wavelet function 'cgau 8', the pixel size of the converted picture is 1054x148, the picture is an RGB color image, and the channel is 3, as shown in FIGS. 2 and 3;
4) building a neural network model combining CNN and LSTM, as shown in figure 4,
the process of building the neural network model combining the CNN and the LSTM comprises the following steps:
4.1 using the time-frequency characteristic energy graph with the uniform format size of 1054x148x3 as the input layer data;
4.2 build up the convolution layer and the pooling layer, and the specific parameters are as follows:
a first layer: the number of filters is 30, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a second layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
and a third layer: the number of filters is 60, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a fourth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
and a fifth layer: the number of filters is 90, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a sixth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
a seventh layer: the number of filters is 120, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
an eighth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
a ninth layer: the number of filters is 150, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a tenth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
wherein the activation function in each convolutional layer is a ReLU activation function;
4.3 build the full connection layer and the LSTM layer, the concrete parameters are: the eleventh layer and the twelfth layer are full connection layers, the number of the neurons is 500 and 100 respectively, and the activation functions are ReLU activation functions; the thirteenth layer is an LSTM layer, where output _ dim is 50 in size; the fourteenth layer is a full connection layer, the number of the neurons is 80, and the activation function is a Softmax activation function;
5) sending the picture data of the training group in the step 3) into a neural network model combining CNN and LSTM for training, in the example, sending all pictures of the training group into the neural network model combining CNN and LSTM in the step 4) for training, and storing each parameter model adjusted by training;
6) classifying and identifying the test group picture data in the step 3) by using the neural network model combining the CNN and the LSTM trained in the step 5), automatically memorizing and counting the result of the classification and identification by using the neural network model combining the CNN and the LSTM, so as to calculate the numerical values of evaluation indexes such as the accuracy, the loss function value and the like of the neural network model combining the CNN and the LSTM, and further evaluating the classification effect of the neural network model combining the CNN and the LSTM by using the numerical values.
As shown in fig. 5, the neural network model combining the CNN and the LSTM trained in step 5) is subjected to simulation testing, in this example, when the number of selected iterations is greater than 7, the classification accuracy of the neural network model combining the CNN and the LSTM is substantially stable, and the accuracy of the classification of the photoplethysmographic signals of different people is high.
Claims (5)
1. An identity recognition method based on photoplethysmography is characterized by comprising the following steps:
1) obtaining training group data and test group data: collecting photoplethysmographic signals of n persons in a specified time period to form a training group; randomly acquiring photoplethysmographic signals of about 3/10 of the n persons in another time period to form a test group, wherein the photoplethysmographic signal data of the training group and the test group are the identities of each person;
2) dividing all photoelectric volume pulse wave signal data in training group data and test group data into a plurality of segments containing a plurality of photoelectric volume pulse waves;
3) converting all the segmented segments into a time-frequency characteristic energy graph form by using continuous wavelet transform;
4) building a neural network model combining CNN and LSTM;
5) sending the picture data of the training group in the step 3) into a neural network model combining CNN and LSTM for training;
6) classifying the test group picture data in the step 3) by using the neural network model combining the CNN and the LSTM trained in the step 5), wherein the classification result is an identity recognition result, and finally evaluating the neural network model combining the CNN and the LSTM trained.
2. The photoplethysmography-based identity recognition method according to claim 1, wherein the number of the test group in step 1) is 2/10 to 4/10 of the total number of the participants.
3. The method for identifying an identity based on photoplethysmography according to claim 1, wherein the segment length of the photoplethysmography in step 2) ranges from 5s to 20s, wherein the number of sampling points in the segment of the photoplethysmography is equal to the product of the sampling frequency and the length of the segmented photoplethysmography segment, and the time interval of the sampling points is 1 divided by the sampling frequency.
4. The photoplethysmography-based identity recognition method according to claim 1, wherein the mother wavelet function of the continuous wavelet transform in step 3) selects any one of 'cgau 8', 'haar', 'dB 2', 'bior' and 'mor 1', the pixel size of the transformed picture is 1054x148, and the picture is an RGB color image with a channel of 3.
5. The identity recognition method based on photoplethysmography according to claim 1, wherein the process of constructing the neural network model combining CNN and LSTM in step 4) is as follows:
4.1, using the time-frequency characteristic energy graph with the uniform format as input layer data;
4.2 build up the convolution layer and the pooling layer, and the specific parameters are as follows:
a first layer: the number of filters is 30, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a second layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
and a third layer: the number of filters is 60, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a fourth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
and a fifth layer: the number of filters is 90, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a sixth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
a seventh layer: the number of filters is 120, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
an eighth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
a ninth layer: the number of filters is 150, the size of kernel _ size is 3x3, and the size of strands is 1x 1;
a tenth layer: max pooling layer, pool _ size 2x2, crosses size 2x 2;
wherein the activation function in each convolutional layer is a ReLU activation function;
4.3 build the full connection layer and the LSTM layer, the concrete parameters are: the eleventh layer and the twelfth layer are full connection layers, the number of the neurons is 500 and 100 respectively, and the activation functions are ReLU activation functions; the thirteenth layer is an LSTM layer, where output _ dim is 50 in size; the fourteenth layer is a full connection layer, the number of the neurons is n, n is the number of the training groups, and the activation function is a Softmax activation function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010194774.8A CN111222498A (en) | 2020-03-19 | 2020-03-19 | Identity recognition method based on photoplethysmography |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010194774.8A CN111222498A (en) | 2020-03-19 | 2020-03-19 | Identity recognition method based on photoplethysmography |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111222498A true CN111222498A (en) | 2020-06-02 |
Family
ID=70830150
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010194774.8A Pending CN111222498A (en) | 2020-03-19 | 2020-03-19 | Identity recognition method based on photoplethysmography |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111222498A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112613431A (en) * | 2020-12-28 | 2021-04-06 | 中北大学 | Automatic identification method, system and device for leaked gas |
| CN113569685A (en) * | 2021-07-20 | 2021-10-29 | 华中科技大学 | Method and system for establishing fault diagnosis model and fault diagnosis of machine tool spindle bearing |
| CN113892919A (en) * | 2021-12-09 | 2022-01-07 | 季华实验室 | Pulse feeling data acquisition method and device, electronic equipment and system |
| CN114098691A (en) * | 2022-01-26 | 2022-03-01 | 之江实验室 | Pulse wave identity authentication method, device and medium based on Gaussian mixture model |
| CN115830656A (en) * | 2022-12-08 | 2023-03-21 | 辽宁科技大学 | Identity recognition method and device based on pulse wave |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110077484A1 (en) * | 2009-09-30 | 2011-03-31 | Nellcor Puritan Bennett Ireland | Systems And Methods For Identifying Non-Corrupted Signal Segments For Use In Determining Physiological Parameters |
| CN106473750A (en) * | 2016-10-08 | 2017-03-08 | 西安电子科技大学 | Personal identification method based on photoplethysmographic optimal period waveform |
| CN107088069A (en) * | 2017-03-29 | 2017-08-25 | 西安电子科技大学 | Personal identification method based on human body PPG signal subsections |
| WO2018145377A1 (en) * | 2017-02-07 | 2018-08-16 | 华为技术有限公司 | User identity recognition method, apparatus and system |
| US20190133468A1 (en) * | 2017-11-03 | 2019-05-09 | Samsung Electronics Co., Ltd. | Method and apparatus for high accuracy photoplethysmogram based atrial fibrillation detection using wearable device |
| CN110288018A (en) * | 2019-06-24 | 2019-09-27 | 桂林电子科技大学 | A kind of WiFi personal identification method merging deep learning model |
| US20190313915A1 (en) * | 2015-06-14 | 2019-10-17 | Facense Ltd. | Posture-adjusted calculation of physiological signals |
| CN110458197A (en) * | 2019-07-11 | 2019-11-15 | 启东市知微电子科技有限公司 | Personal identification method and its system based on photoplethysmographic |
-
2020
- 2020-03-19 CN CN202010194774.8A patent/CN111222498A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110077484A1 (en) * | 2009-09-30 | 2011-03-31 | Nellcor Puritan Bennett Ireland | Systems And Methods For Identifying Non-Corrupted Signal Segments For Use In Determining Physiological Parameters |
| US20190313915A1 (en) * | 2015-06-14 | 2019-10-17 | Facense Ltd. | Posture-adjusted calculation of physiological signals |
| CN106473750A (en) * | 2016-10-08 | 2017-03-08 | 西安电子科技大学 | Personal identification method based on photoplethysmographic optimal period waveform |
| WO2018145377A1 (en) * | 2017-02-07 | 2018-08-16 | 华为技术有限公司 | User identity recognition method, apparatus and system |
| CN107088069A (en) * | 2017-03-29 | 2017-08-25 | 西安电子科技大学 | Personal identification method based on human body PPG signal subsections |
| US20190133468A1 (en) * | 2017-11-03 | 2019-05-09 | Samsung Electronics Co., Ltd. | Method and apparatus for high accuracy photoplethysmogram based atrial fibrillation detection using wearable device |
| CN110288018A (en) * | 2019-06-24 | 2019-09-27 | 桂林电子科技大学 | A kind of WiFi personal identification method merging deep learning model |
| CN110458197A (en) * | 2019-07-11 | 2019-11-15 | 启东市知微电子科技有限公司 | Personal identification method and its system based on photoplethysmographic |
Non-Patent Citations (1)
| Title |
|---|
| 陈真诚 等: "基于光电容积脉搏波信号与心电信号相关性的研究", 《生物医学工程研》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112613431A (en) * | 2020-12-28 | 2021-04-06 | 中北大学 | Automatic identification method, system and device for leaked gas |
| CN112613431B (en) * | 2020-12-28 | 2021-06-29 | 中北大学 | Automatic identification method, system and device for leaked gas |
| CN113569685A (en) * | 2021-07-20 | 2021-10-29 | 华中科技大学 | Method and system for establishing fault diagnosis model and fault diagnosis of machine tool spindle bearing |
| CN113892919A (en) * | 2021-12-09 | 2022-01-07 | 季华实验室 | Pulse feeling data acquisition method and device, electronic equipment and system |
| CN113892919B (en) * | 2021-12-09 | 2022-04-22 | 季华实验室 | Pulse feeling data acquisition method and device, electronic equipment and system |
| CN114098691A (en) * | 2022-01-26 | 2022-03-01 | 之江实验室 | Pulse wave identity authentication method, device and medium based on Gaussian mixture model |
| CN115830656A (en) * | 2022-12-08 | 2023-03-21 | 辽宁科技大学 | Identity recognition method and device based on pulse wave |
| CN115830656B (en) * | 2022-12-08 | 2023-07-14 | 辽宁科技大学 | Pulse wave-based identity recognition method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111222498A (en) | Identity recognition method based on photoplethysmography | |
| Yuan et al. | Fingerprint liveness detection using an improved CNN with image scale equalization | |
| CN110069958B (en) | Electroencephalogram signal rapid identification method of dense deep convolutional neural network | |
| CN108776788B (en) | Brain wave-based identification method | |
| CN109784023B (en) | Steady-state vision-evoked electroencephalogram identity recognition method and system based on deep learning | |
| CN110458197A (en) | Personal identification method and its system based on photoplethysmographic | |
| CN110353702A (en) | A kind of emotion identification method and system based on shallow-layer convolutional neural networks | |
| CN111523601B (en) | Potential emotion recognition method based on knowledge guidance and generation of countermeasure learning | |
| CN109497990B (en) | Electrocardiosignal identity recognition method and system based on canonical correlation analysis | |
| Huang et al. | Human identification with electroencephalogram (EEG) signal processing | |
| Zhou et al. | Research on image preprocessing algorithm and deep learning of iris recognition | |
| Patil et al. | A non-contact PPG biometric system based on deep neural network | |
| CN111797747A (en) | Potential emotion recognition method based on EEG, BVP and micro-expression | |
| CN109620260A (en) | Psychological condition recognition methods, equipment and storage medium | |
| CN108256579A (en) | A kind of multi-modal sense of national identity quantization measuring method based on priori | |
| CN113243924A (en) | Identity recognition method based on electroencephalogram signal channel attention convolution neural network | |
| CN115969392A (en) | Cross-period brainprint recognition method based on tensor frequency space attention domain adaptive network | |
| CN114081505A (en) | Electroencephalogram signal identification method based on Pearson correlation coefficient and convolutional neural network | |
| CN116861217A (en) | Identity recognition method and system for mobile terminal | |
| He et al. | Emotion classification using eeg data in a brain-inspired spiking neural network | |
| Krishna et al. | Ear-Based Biometric System Using Artificial Intelligence | |
| Nawas et al. | K-NN classification of brain dominance | |
| Cheng et al. | Heart sound recognition-A prospective candidate for biometric identification | |
| Li et al. | Authentication study for brain-based computer interfaces using music stimulations | |
| CN113951886A (en) | Brain magnetic pattern generation system and lie detection decision system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200602 |
|
| RJ01 | Rejection of invention patent application after publication |