Heart rate detection method and apparatus, storage medium, and electronic device

Abstract

The application discloses a heart rate detection method and device, a storage medium and an electronic device. A pulse wave signal and an acceleration signal of a user are acquired, and a current state of the user is determined. When the user is in a static state, time domain analysis is performed based on the pulse wave signal to obtain at least one first static heart rate value, and frequency domain analysis is performed based on the pulse wave signal to obtain a plurality of second static heart rate values. When the user is in a motion state, frequency domain analysis is performed based on the pulse wave signal to obtain a plurality of first motion heart rate values, and deep learning analysis is performed based on the pulse wave signal and the acceleration signal to obtain a second motion heart rate value. A target heart rate value of the user is determined according to the first static heart rate value and the second static heart rate value, or the target heart rate value of the user is determined according to the first motion heart rate value and the second motion heart rate value. In the static state, the time domain and the frequency domain are combined, the accurate heart rate value is obtained, and the energy consumption is reduced. In the motion state, the frequency domain and the deep learning are combined, and the more accurate heart rate value is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wearable devices, and in particular to a heart rate detection method and device, a storage medium, and an electronic device. BACKGROUND

[0002] With the continuous development of optical sensors and advanced signal processing algorithms, smart wearable devices, such as sports watches, can achieve heart rate monitoring of users through photoplethysmography (PPG). It is based on optical sensors, which measure the degree of attenuation of light reflected by blood vessels and other tissues on the surface of the human body, record the pulsation state of blood vessels, measure the pulse wave, and thus correlate the beating of the heart and measure the heart rate. However, in the case of a wearer performing strenuous exercise or a large environmental disturbance, the pulse wave signal can be affected by noise, resulting in inaccurate heart rate measurement. SUMMARY

[0003] The present application provides a heart rate detection method, device, storage medium, and electronic device to solve the technical problem of low accuracy of a single pulse wave signal in heart rate detection.

[0004] In a first aspect, an embodiment of the present application provides a heart rate detection method, which comprises:

[0005] Obtaining a pulse wave signal and an acceleration signal of a user, determining a current state of the user, and the current state comprising a static state or a motion state;

[0006] When the user is in the static state, performing time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, and performing frequency domain analysis based on the pulse wave signal to obtain a plurality of second static heart rate values;

[0007] When the user is in the motion state, performing frequency domain analysis based on the pulse wave signal to obtain a plurality of first motion heart rate values, and performing deep learning analysis based on the pulse wave signal and the acceleration signal to obtain a second motion heart rate value;

[0008] Determining a target heart rate value of the user according to the first static heart rate values and the second static heart rate values, or determining a target heart rate value of the user according to the first motion heart rate values and the second motion heart rate value.

[0009] In a second aspect, an embodiment of the present application provides a heart rate detection device, which comprises:

[0010] A state determination unit configured to obtain a pulse wave signal and an acceleration signal of a user, determine a current state of the user, and the current state comprising a static state or a motion state;

[0011] The first calculation unit is configured to, when the user is in a static state, perform time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, and perform frequency domain analysis based on the pulse wave signal to obtain a plurality of second static heart rate values;

[0012] The second calculation unit is configured to, when the user is in a dynamic state, perform frequency domain analysis based on the pulse wave signal to obtain a plurality of first dynamic heart rate values, and perform deep learning analysis based on the pulse wave signal and the acceleration signal to obtain a second dynamic heart rate value.

[0013] The result determination unit is configured to determine a target heart rate value of the user according to the first static heart rate values and the second static heart rate values, or determine the target heart rate value of the user according to the first dynamic heart rate values and the second dynamic heart rate value.

[0014] In a third aspect, an embodiment of the present application provides a computer storage medium, which stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and performing the steps of the method.

[0015] In a fourth aspect, an embodiment of the present application provides an electronic device, which comprises a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the computer program is suitable for being loaded by the processor and performing the steps of the method.

[0016] The technical scheme provided by some embodiments of the present application has at least the following beneficial effects:

[0017] The application provides a heart rate detection method, which acquires a pulse wave signal and an acceleration signal of a user, judges a current state of the user, and the current state includes a static state or a motion state; when the user is in the static state, time domain analysis is performed based on the pulse wave signal to obtain at least one first static heart rate value, and frequency domain analysis is performed based on the pulse wave signal to obtain a plurality of second static heart rate values; when the user is in the motion state, frequency domain analysis is performed based on the pulse wave signal to obtain a plurality of first motion heart rate values, and deep learning analysis is performed based on the pulse wave signal and the acceleration signal to obtain a second motion heart rate value; a target heart rate value of the user is determined according to the first static heart rate values and the second static heart rate values, or the target heart rate value of the user is determined according to the first motion heart rate values and the second motion heart rate value. The pulse wave signal and the acceleration signal are collected, which provides a data basis for subsequent state judgment and heart rate analysis. The acceleration signal can effectively assist in distinguishing the pulse wave change caused by the motion from the real heart rate change, which is helpful for more comprehensively understanding the heart rate change of the user. Meanwhile, accurately judging the current state of the user is a prerequisite for subsequently selecting an appropriate heart rate detection method, which can ensure that the time domain analysis and the frequency domain analysis with reduced power consumption are adopted in the static state, and the frequency domain analysis and the deep learning analysis are more effectively utilized in the motion state to reduce the influence of the motion artifact. When the user is in the static state, the time domain analysis based on the pulse wave signal can directly reflect the change trend of the pulse wave to obtain the basic information of the heart rate, and meanwhile, the pulse wave signal is converted from the time domain to the frequency domain to observe the distribution of the pulse wave at different frequencies, so that a plurality of heart rate values in the static state are obtained. When the user is in the motion state, in addition to the frequency domain analysis, a deep learning model is introduced to comprehensively analyze the pulse wave signal and the acceleration signal. The deep learning model can learn and identify the pulse wave change mode caused by the motion, so as to effectively reduce the influence of the motion artifact on the heart rate detection and more accurately extract the real heart rate information in the motion state. Finally, the target heart rate value of the user is determined according to the heart rate value in the static state or the heart rate value in the motion state, which can help the user understand the heart rate condition of the user and serve as an important basis for health management and disease prevention. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, below the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0019] Figure 1 An exemplary system architecture diagram of a heart rate detection method provided by the embodiments of the present application is shown in the figure.

[0020] Figure 2A flowchart of a heart rate detection method provided by an embodiment of the present application is shown in FIG. 1.

[0021] Figure 3 A general flowchart of a heart rate detection method provided by an embodiment of the present application is shown in FIG. 2.

[0022] Figure 4 A flowchart of a heart rate detection method provided by an embodiment of the present application is shown in FIG. 3.

[0023] Figure 5 A flowchart of a time domain analysis in a heart rate detection method provided by an embodiment of the present application is shown in FIG. 4.

[0024] Figure 6 A flowchart of a heart rate detection method provided by an embodiment of the present application is shown in FIG. 5.

[0025] Figure 7 A flowchart of a heart rate detection method provided by an embodiment of the present application is shown in FIG. 6.

[0026] Figure 8 A flowchart of a frequency domain analysis in a heart rate detection method provided by an embodiment of the present application is shown in FIG. 7.

[0027] Figure 9 A flowchart of a deep learning analysis in a heart rate detection method provided by an embodiment of the present application is shown in FIG. 8.

[0028] Figure 10 A flowchart of a heart rate detection method provided by an embodiment of the present application is shown in FIG. 9.

[0029] Figure 11 A structural block diagram of a heart rate detection device provided by an embodiment of the present application is shown in FIG. 10.

[0030] Figure 12 A structural diagram of an electronic device provided by an embodiment of the present application is shown in FIG. 11. DETAILED DESCRIPTION

[0031] In order to make the features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0032] The following description refers to the accompanying drawings. Unless otherwise noted, same or similar components in different drawings have same or similar reference numerals. The implementations described in the following exemplary embodiments are not meant to represent all implementations consistent with the present disclosure. Instead they are simply examples of apparatuses and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0033] Smart wearable devices such as smartwatches can monitor the heart rate of a user through a pulse wave signal (PPG signal). This is based on an optical sensor, by measuring the degree of attenuation of light reflected by blood vessels and other tissues on the surface of the human skin, recording the pulsation state of the blood vessels, measuring the pulse wave, and thus correlating the beating of the heart and measuring the heart rate. This achieves non-invasive and continuous heart rate monitoring, greatly improving the convenience of heart rate monitoring.

[0034] Although the pulse wave signal can provide accurate heart rate monitoring in a stationary state, when the wearer is performing strenuous exercise, the pulse wave signal may be disturbed by noise due to muscle contraction, skin vibration and other factors; in addition, environmental factors such as light changes, temperature changes and the like may also interfere with the pulse wave signal. These factors can cause inaccurate heart rate measurement, or even fail to obtain effective heart rate data, thereby affecting the accuracy of health assessment.

[0035] Therefore, the embodiments of the present application provide a heart rate detection method to solve the technical problem of low accuracy of a single pulse wave signal in heart rate detection.

[0036] Please refer to Figure 1 , Figure 1 An exemplary system architecture diagram of a heart rate detection method provided by the embodiments of the present application.

[0037] As Figure 1 shown, the system architecture can include an electronic device 101, a network 102 and a server 103. The network 102 is used to provide a communication link medium between the electronic device 101 and the server 103. The network 102 can include various types of wireless communication links, such as: Bluetooth communication link, Wireless-Fidelity (Wi-Fi) communication link or microwave communication link, etc.

[0038] The electronic device 101 can interact with the server 103 through the network 102 to receive or send messages from or to the server 103, or the electronic device 101 can interact with the server 103 through the network 102 to receive messages or data sent by other users to the server 103. For example, the electronic device 101 can upload the pulse wave signal and the acceleration signal to the server 103 through the network 102 (such as Bluetooth, Wi-Fi), and then the server 103 processes these data by using a deep learning algorithm and returns multiple exercise heart rate values to the watch to reflect the heart health status of the user during exercise.

[0039] The electronic device 101 can be hardware or software. When the electronic device 101 is hardware, it can be various electronic devices including but not limited to a smart watch, a smart ring, etc. When the electronic device 101 is software, it can be installed in the above-mentioned electronic devices, and can be implemented as multiple software or software modules (for example, used to provide distributed services) or a single software or software module, which is not specifically limited here.

[0040] In the embodiments of the present application, the electronic device 101 first acquires the pulse wave signal and the acceleration signal of the user, judges the current state of the user, and the current state includes a static state or a motion state. Then, when the user is in the static state, the electronic device 101 performs time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, and performs frequency domain analysis based on the pulse wave signal to obtain multiple second static heart rate values. When the user is in the motion state, the electronic device 101 performs frequency domain analysis based on the pulse wave signal to obtain multiple first exercise heart rate values, and performs deep learning analysis based on the pulse wave signal and the acceleration signal to obtain a second exercise heart rate value. Finally, the electronic device 101 determines the target heart rate value of the user according to the first static heart rate values and the second static heart rate values, or determines the target heart rate value of the user according to the first exercise heart rate values and the second exercise heart rate value.

[0041] The server 103 can be a service server providing various services. It should be noted that the server 103 can be hardware or software. When the server 103 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server. When the server 103 is software, it can be implemented as multiple software or software modules (for example, used to provide distributed services), or as a single software or software module, which is not specifically limited here.

[0042] Alternatively, the system architecture can also not include the server 103, in other words, the server 103 can be an optional device in the embodiments of the present specification, that is, the method provided in the embodiments of the present specification can be applied to a system architecture including only the electronic device 101, and the embodiments of the present application do not limit this.

[0043] It should be understood that Figure 1 The number of electronic devices, networks and servers in the above system architecture is only illustrative, and can be any number of electronic devices, networks and servers according to the needs of implementation.

[0044] Please refer to Figure 2 , Figure 2 A flowchart of a heart rate detection method provided in the embodiments of the present application. The execution subject of the embodiments of the present application can be an electronic device that executes heart rate detection, or a processor in an electronic device that executes a heart rate detection method, or a heart rate detection service in an electronic device that executes a heart rate detection method. For the convenience of description, the specific execution process of the heart rate detection method is introduced below taking the execution subject as the processor in the electronic device.

[0045] As Figure 2 indicated, the heart rate detection method can at least include:

[0046] S202, acquiring a pulse wave signal and an acceleration signal of a user, and determining a current state of the user, the current state including a static state or a motion state.

[0047] Optionally, the pulse wave signal is a direct data source for heart rate detection, which provides basic information of heartbeats. When the user is in a static state, the pulse wave signal is easily affected by environmental noise such as light and temperature, and the influence of motion interference factors on the pulse wave signal is small, so the pulse wave signal is relatively stable. When the user is in a motion state, due to the increase of body activity, the pulse wave signal can be affected by motion artifacts, resulting in a decrease in the accuracy of the heart rate detection result. Based on this, different heart rate detection methods can be considered according to different states of the user. When the user is in a static state, the influence of environmental interference factors on the pulse wave signal is considered, and the influence of motion interference factors does not need to be considered, thereby reducing the utilization of computing resources and reducing power consumption; when the user is in a motion state, motion artifacts need to be filtered out to improve the accuracy of heart rate detection.

[0048] Optionally, when the user is in a static state, the activity of each part of the body is reduced, the corresponding acceleration change is relatively small, and a relatively stable feature is presented; when the user starts to move, the activity of each part of the body is increased, resulting in that the acceleration presents a clear change feature, such as sudden increase of the acceleration, frequent change of the direction, and the like. Based on this, the user is in a static state or a moving state can be determined by processing and analyzing the acceleration signal of the user.

[0049] Optionally, when the heart rate is detected by the wearable device, the pulse wave signal and the acceleration signal of the user can be acquired by the sensors, such as the optical volume pulse wave sensor and the acceleration sensor, built in the wearable device (such as the sports watch, the sports ring, and the like), wherein the pulse wave signal is the direct data source of the heart rate detection; the acceleration signal is used as auxiliary information to help determine the state of the user and optimize the heart rate detection method. Specifically, the activity parameters capable of quantifying the activity amount of the user, such as the acceleration mean and variance, the step count, the energy consumption estimation, and the like, can be extracted by processing the acceleration signal. Then, the current state of the user is comprehensively determined according to the activity parameters, for example, when the acceleration mean is close to zero, the variance is very small, and the step count is zero or very low, it can be determined that the user is in a static state; when the acceleration signal shows high mean, large variance, and frequent peaks, and the step count is very high or cannot be accurately counted (such as running, jumping, and the like), it can be determined that the user is in a moving state.

[0050] It should be noted that, in addition to the acceleration signal, the current state of the user can also be determined by other ways, for example, the physiological parameters such as the breathing frequency and the skin conductivity are used to determine whether the user is in a moving state, and the activity history and the like of the user can be combined to make more accurate state determination. The method for determining the state of the user is not limited in the embodiments of the present application.

[0051] Optionally, Figure 3 The overall flowchart of the heart rate detection method provided in the embodiments of the present application is shown in FIG. 1. Figure 3As shown in S302-306, in order to remove noise interference such as environmental electromagnetic interference, baseline drift caused by respiration, and the like, and retain the frequency components related to heart rate, the acquired pulse wave signal and acceleration signal can also be respectively subjected to time domain band-pass filtering processing of the heart rate frequency band to obtain a filtered pulse wave band-pass signal and an acceleration band-pass signal, and the user state is determined and the subsequent heart rate detection steps are performed according to the pulse wave band-pass signal and the acceleration band-pass signal. Specifically, the filter parameters can be selected according to the preset heart rate frequency band (usually the adult heart rate range is 35-240 times / minute, and the corresponding frequency band is about 0.58-4Hz). Here, at least one method can be used among Finite Impulse Response Filter (FIR), Infinite Impulse Response Filter (IIR), and Discrete Wavelet Transform (DWT) to remove the noise (for example, the noise of 0.1 Hz in the heart rate frequency band) and interference components in the signal and retain the effective signal related to the heart rate.

[0052] S204, when the user is in a static state, time domain analysis is performed based on the pulse wave signal to obtain at least one first static heart rate value, and frequency domain analysis is performed based on the pulse wave signal to obtain a plurality of second static heart rate values.

[0053] Optionally, since when the user is in a static state, the influence of motion interference factors on the pulse wave signal is small, and the pulse wave signal is relatively stable, when selecting the heart rate detection method in the static state, the power consumption can be saved as much as possible while ensuring the accuracy of the result. Based on this, as shown in S206, when the user is in a static state, the heart rate detection method based on the pulse wave signal can be selected from the following three methods: Figure 3 As shown in S308, in the static state, the time domain analysis and the frequency domain analysis can be combined to fully utilize the different dimensional information of the signal, to mutually check and supplement each other, to improve the accuracy of heart rate detection, and to maintain a low power consumption level.

[0054] Optionally, the time domain analysis is to regard the signal as a function of time, which is intuitive and can directly show the changes of the signal over time, such as the waveform, amplitude, phase, etc. of the signal, and is sensitive to the instantaneous characteristics of the signal, and can capture the small changes of the signal. Moreover, since the time domain analysis mainly focuses on the changes of the signal over time, the calculation and processing process is relatively simple, and does not require complex calculation or large amount of data conversion, so as to reduce the power consumption of the processor. The frequency domain analysis is to convert the signal from the time domain to the frequency domain by Fourier transform or the like, analyze the spectral characteristics of the signal, and further extract the heart rate information. This process regards the signal as a function of frequency, and analyzes the characteristics of the signal by decomposition and reconstruction of the signal on the frequency axis. The frequency domain analysis can accurately identify the components and distribution of the signal at different frequencies, and provide deeper heart rate information. Therefore, in the stationary state, the pulse wave signal is relatively stable, the time domain analysis can directly show the waveform and timing relationship of the signal, and the frequency domain analysis can reveal the frequency components of the signal, and the combination of the two can more comprehensively understand the characteristics of the heart rate.

[0055] Specifically, in the time domain analysis, the wave peak or wave trough of the pulse wave signal (or the pulse wave bandpass signal) can be identified by using the wave searching algorithm, the pulse period between adjacent wave peaks (or wave troughs) is calculated, and then a plurality of first stationary heart rate values are calculated according to the pulse period. In addition, since the heart rate is relatively stable in the stationary state, the heart rate values of a plurality of consecutive periods can also be calculated, and the average value is taken as one of the stationary heart rate values of this stage to reduce accidental errors. In the frequency domain analysis, the fast Fourier transform (FFT) can be performed on the pulse wave signal (or the pulse wave bandpass signal) to obtain the frequency spectrum diagram of the signal. Then in the frequency spectrum diagram, a plurality of spectral components corresponding to the heart rate frequency are searched, which are usually represented as obvious peaks, and the frequency corresponding to the peaks is the heart rate frequency, and the heart rate frequencies are converted into second stationary heart rate values. The first stationary heart rate values obtained by the time domain analysis and the second stationary heart rate values obtained by the frequency domain analysis are the stationary heart rate values in the stationary state.

[0056] S206, when the user is in a motion state, performing frequency domain analysis based on the pulse wave signal to obtain a plurality of first motion heart rate values, and performing deep learning analysis based on the pulse wave signal and the acceleration signal to obtain a second motion heart rate value.

[0057] Optionally, in the motion state, there are more motion interference signals generated by body movement, so when selecting the heart rate detection method in the motion state, the motion artifacts need to be considered for filtering to improve the accuracy of heart rate detection. Based on this, as Figure 3In the stationary state, the pulse wave signal and the acceleration signal can be processed in combination with frequency domain analysis and deep learning analysis, as shown in S310, to cope with complex motion interference and heart rate changes and provide more accurate heart rate values.

[0058] Optionally, the heart rate changes greatly in the motion state, and the frequency domain analysis can identify and process such changes. The pulse-type interference in the time domain signal has good anti-interference ability, and can more accurately extract heart rate information. The deep learning algorithm is good at processing complex nonlinear signals and sequence data, can automatically learn and extract key features in the signal, and capture the dynamic changes of the pulse wave signal in the motion state, thereby improving the accuracy of heart rate detection. In addition, in the motion state, the acceleration signal can reflect the motion state and motion intensity of the user, providing additional information for heart rate detection and further improving the accuracy of heart rate detection.

[0059] Specifically, in the deep learning analysis in the motion state, a large amount of labeled data (such as electrocardiogram, pulse wave signal, acceleration signal, and true heart rate value recorded at the same time) can be used to train the model in the training stage, and the model can learn the mapping relationship from the input signal to the heart rate value by optimizing the loss function. In the actual heart rate detection process, the preprocessed pulse wave signal and acceleration signal can be selected to be input into the trained deep learning model, and the second motion heart rate value can be obtained by multiple predictions or averaging of the input signal by the model. Then the first motion heart rate value obtained by frequency domain analysis and the second motion heart rate value obtained by deep learning analysis are used as the motion heart rate value in the motion state.

[0060] It should be noted that the selection of the deep learning model can be a neural network regression model such as a convolutional neural network, a recurrent neural network, or other models, which is selected according to actual needs. For example, in a wearable device such as a sports watch or a sports ring that needs to consider power consumption, a neural network regression model with relatively low power consumption can be selected.

[0061] S208, determining the target heart rate value of the user according to the first stationary heart rate values and the second stationary heart rate values, or determining the target heart rate value of the user according to the first motion heart rate values and the second motion heart rate values.

[0062] Optionally, as Figure 3 As shown in S312, after the multiple stationary heart rate values or motion heart rate values are preliminarily determined, the most accurate and representative heart rate value needs to be selected from the multiple heart rate values as the target heart rate value of the user for output.

[0063] Optionally, first, obviously abnormal or unreasonable data points caused by signal interference, measurement error, etc. are removed. Next, a weighted average method can be used to determine the target heart rate value. Specifically, a weight can be assigned to each heart rate value, and the determination of the weight can be based on various factors, such as signal quality, effectiveness of the analysis method, consistency with other heart rate values, accuracy of historical data, etc. For example, when the acceleration signal shows that the user is in a motion state, more reliance can be placed on the heart rate value obtained by deep learning analysis, and a higher weight is assigned to the motion heart rate value obtained by deep learning analysis, so that the target heart rate value is closer to the user's true heart rate. Then, the target heart rate value is output to the user or the relevant system in an appropriate form, so that the user can understand his own health status. It should be noted that in addition to the weighted average, other methods such as median and mode can also be used to determine the target heart rate value, and the specific method of obtaining the target heart rate value is not limited in the embodiments of the present application.

[0064] In this embodiment, a heart rate detection method is provided, which acquires a user's pulse wave signal and acceleration signal to determine the user's current state, which includes a resting state or a moving state. When the user is in a resting state, time-domain analysis is performed based on the pulse wave signal to obtain at least one first resting heart rate value, and frequency-domain analysis is performed based on the pulse wave signal to obtain multiple second resting heart rate values. When the user is in a moving state, frequency-domain analysis is performed based on the pulse wave signal to obtain multiple first moving heart rate values, and deep learning analysis is performed based on the pulse wave signal and acceleration signal to obtain second moving heart rate values. The user's target heart rate value is determined based on each first resting heart rate value and each second resting heart rate value, or the user's target heart rate value is determined based on each first moving heart rate value and each second moving heart rate value. By collecting pulse wave and acceleration signals, a data foundation is provided for subsequent state assessment and heart rate analysis. Acceleration signals effectively help distinguish between pulse wave changes caused by exercise and actual heart rate changes, contributing to a more comprehensive understanding of the user's heart rate fluctuations. Accurately determining the user's current state is a prerequisite for selecting an appropriate heart rate detection method. This ensures that time-domain and frequency-domain analysis with reduced power consumption are used in a static state, while frequency-domain analysis and deep learning analysis are more effectively utilized to reduce the impact of motion artifacts during exercise. When the user is at rest, time-domain analysis based on the pulse wave signal can directly reflect the trend of pulse wave changes, obtaining basic heart rate information. The pulse wave signal is converted from the time domain to the frequency domain to observe its distribution at different frequencies, thus obtaining multiple resting heart rate values. When the user is in motion, in addition to frequency domain analysis, a deep learning model is introduced to comprehensively analyze the pulse wave signal and acceleration signal. The deep learning model can learn and identify pulse wave change patterns caused by motion, thereby effectively reducing the impact of motion artifacts on heart rate detection and more accurately extracting the true heart rate information during exercise. Finally, the user's target heart rate value is determined based on the resting or exercising heart rate value, which helps the user understand their heart rate status and serves as an important basis for health management and disease prevention.

[0065] Please see Figure 4 , Figure 4 This is a flowchart illustrating a heart rate detection method provided in an embodiment of this application.

[0066] like Figure 4 As shown, heart rate detection methods can include at least:

[0067] S402. Acquire the user's pulse wave signal and acceleration signal, and determine the user's current state, which may include a stationary state or a moving state.

[0068] Optionally, as to step S402, please refer to the detailed description in step S202, which will not be repeated here.

[0069] S404, time domain band-pass filtering the pulse wave signal according to a preset heart rate frequency band to obtain a pulse wave band-pass signal; when the user is in a static state, performing peak-valley searching and / or establishing a probability density function on the pulse wave band-pass signal to obtain at least one first static heart rate value.

[0070] Optionally, since the pulse wave signal can contain multiple frequency components, including components related to heart rate and other interference components (such as signals generated by breathing frequency, light influence, device noise, etc.). Therefore, after obtaining the pulse wave signal, the pulse wave signal can be time domain band-pass filtered to obtain a pulse wave band-pass signal to remove irrelevant noise components and obtain a more pure pulse wave band-pass signal, and based on which subsequent heart rate value acquisition is performed.

[0071] Optionally, in time domain analysis, the pulse wave signal is represented with time as the horizontal axis and pulse wave amplitude (heart beat amplitude) as the vertical axis, which directly reflects the change of the pulse wave generated by the heart with time each time the heart beats. In a static state, due to the periodic beating of the heart, the pulse wave signal will show a periodic fluctuation, in which the peak usually corresponds to the systole period of the heart, and the valley corresponds to the diastole period of the heart, and the time interval between the peaks or valleys reflects the period of the heart beat, i.e. the heart rate. Based on this, Figure 5 A flowchart of a time domain analysis method provided by an embodiment of the present application is shown in Figure 5 In the time domain analysis method in a static state, as shown in S502-S504, the filtered pulse wave band-pass signal can be first obtained, and peak-valley searching can be performed on the pulse wave band-pass signal. This can identify the peak and valley positions in the signal by setting appropriate threshold values and searching algorithms, and calculate a plurality of corresponding first static heart rate values according to the time interval between the peaks or valleys in different periods. In addition, there can still be some small fluctuations or abnormal values in the filtered pulse wave band-pass signal. Considering the periodic characteristics of the signal, we can obtain a more stable first static heart rate value by calculating the average time interval between a plurality of peaks or valleys, further reducing the influence of random errors and noise interference. At the same time, other statistical methods (such as median, mode, etc.) can also be used to process a plurality of heart rate values to obtain more reliable results.

[0072] Optionally, the probability density function is used to describe the probability distribution followed by a continuous random variable. In signal analysis, since the intervals between peaks and troughs can take any real value (within a certain range), they can be considered as continuous random variables. Therefore, we can use the probability density function to describe the probability distribution of these intervals, thus intuitively showing the distribution characteristics of the peak or trough intervals in the signal. Based on this, as... Figure 5 As shown in S506, the probability density function of the pulse wave bandpass signal can be used to represent the probability values ​​of different heart rates, where the peak position usually corresponds to the most likely first heart rate value. Furthermore, one or more probability thresholds can be set according to actual needs, and heart rate values ​​that meet these thresholds can be used as possible first resting heart rate values. In some cases, the probability density function may exhibit multiple peaks, and multiple heart rate values ​​may meet the probability thresholds. These heart rate values ​​can be initially identified as the first resting heart rate values, and further screening and determination of the target heart rate value can be achieved by combining other analytical methods (such as frequency domain analysis).

[0073] S406. Differentiate the pulse wave bandpass signal to obtain the pulse wave differential signal; perform peak and trough search and / or establish a probability density function on the pulse wave differential signal to obtain at least one first resting heart rate value.

[0074] Optionally, such as Figure 5 As shown in S502-508, to further improve the accuracy of heart rate value acquisition and reduce the impact of low-frequency interference such as baseline drift and light changes, differential processing can be performed on the pulse wave bandpass signal. Specifically, the bandpass signal is differentially processed sequentially, and the difference between adjacent sampling points is used as the new signal value to obtain the pulse wave differential signal. In this way, low-frequency interference is reduced in the differential signal, while feature points directly related to heart rate, such as peaks and troughs, are enhanced, making them easier to identify and extract in subsequent analysis. Next, peak and trough search and probability density function are also performed on the pulse wave differential signal to obtain the possible first resting heart rate value. For details on how to perform peak and trough search and establish probability density function on the pulse wave differential signal, please refer to the detailed description in step S404, which will not be repeated here.

[0075] S408, and based on the pulse wave signal, frequency domain analysis was performed to obtain multiple second resting heart rate values.

[0076] Optionally, in addition to time-domain analysis, frequency-domain analysis can also be performed on the pulse wave bandpass signal in the resting state. For details regarding step S408, please refer to the detailed description in step S204; it will not be repeated here.

[0077] S410, when the user is in a motion state, performing frequency domain analysis based on the pulse wave signal to obtain a plurality of first motion heart rate values, and performing deep learning analysis based on the pulse wave signal and the acceleration signal to obtain a second motion heart rate value.

[0078] Optionally, for step S410, please refer to the detailed description in step S206, which will not be repeated here.

[0079] S412, determining a target heart rate value of the user according to the first and second static heart rate values, or determining a target heart rate value of the user according to the first and second motion heart rate values.

[0080] Optionally, for step S412, please refer to the detailed description in step S208, which will not be repeated here.

[0081] In the embodiments of the present application, a heart rate detection method is provided, which performs peak and valley searching on the pulse wave band communication signal, can calculate the first static heart rate value through the key feature points of the peaks and valleys which directly reflect the heart beat period, and further intuitively obtain the possible first static heart rate value by establishing a probability density function of the pulse wave band communication signal to describe the probability distribution of different heart rates, thereby reducing the use of computing resources and reducing power consumption; further, the pulse wave band communication signal is subjected to differential processing, and the differential pulse wave differential signal is subjected to peak and valley searching and establishment of a probability density function, which can enhance the change components in the signal, thereby removing the low-frequency interference components in the pulse wave signal, and help to more accurately identify and extract the feature points directly related to the heart rate, and further enhance the accuracy of heart rate detection.

[0082] Please refer to Figure 6 , Figure 6 A flowchart of a heart rate detection method provided in the embodiments of the present application is shown.

[0083] As Figure 6 shown, the heart rate detection method can at least include:

[0084] S602, obtaining a pulse wave signal and an acceleration signal of a user, performing time domain band pass filtering on the pulse wave signal according to a preset heart rate frequency band to obtain a pulse wave band communication signal, performing time domain band pass filtering on the acceleration signal according to the preset heart rate frequency band to obtain an acceleration band communication signal, determining an activity parameter of the user according to the acceleration band communication signal, determining that the user is in a static state when the activity parameter is less than a preset threshold, and determining that the user is in a motion state when the activity parameter is greater than or equal to the preset threshold.

[0085] Optionally, firstly, the pulse wave signal and the acceleration signal of the user are acquired by the sensors built-in the wearable device respectively, then the pulse wave signal is time domain band-pass filtered to obtain a pulse wave band-pass signal; similarly, the multi-channel acceleration signals can also be time domain band-pass filtered in the heart rate frequency band to obtain multi-channel acceleration band-pass signals (for example, x, y, z three-channel acceleration band-pass signals). Specifically, how to time domain band-pass filter the signal in the heart rate frequency band is described in detail in step S202, which will not be repeated here. Further, the current state of the user is determined according to the filtered acceleration band-pass signal.

[0086] Specifically, the acceleration band-pass signal is processed to extract activity parameters that can quantify the activity amount of the user, such as acceleration mean and variance, step count, energy consumption estimation, etc. The above activity parameters are comprehensively analyzed, and the current state of the user is determined in combination with the preset threshold or model. For example, when the activity parameter is lower than a certain threshold, it can be considered that the user is in a static state; when the activity parameter is higher than the threshold and meets the motion characteristics, it can be considered that the user is in a motion state.

[0087] Optionally, a three-axis acceleration sensor is used to measure the acceleration signal of the user, and outputs the acceleration signals in x, y, and z directions respectively. The three-axis acceleration sensor is a sensor that can measure the acceleration of the user in three orthogonal directions simultaneously, which can provide the complete acceleration vector of the user in three-dimensional space, thereby more accurately reflecting the motion state of the user in three-dimensional space. Among them, the static state is usually manifested as the values of the acceleration signals in each direction are close to zero, and the motion state is manifested as the acceleration signals in at least one direction change significantly.

[0088] S604, when the user is in a static state, time domain analysis is performed based on the pulse wave signal to obtain at least one first static heart rate value.

[0089] Optionally, for step S604, please refer to the detailed description in step S204, which will not be repeated here.

[0090] S606, and the first frequency spectrum corresponding to the pulse wave band-pass signal is acquired, and the first frequency spectrum is searched for the heart rate value to obtain a plurality of second static heart rate values.

[0091] Optionally, when frequency domain analysis based on the pulse wave signal is required in the static state, the pulse wave bandpass signal is first obtained, and then the fast Fourier transform is performed on the pulse wave bandpass signal to obtain its frequency spectrum representation, i.e., the first frequency spectrum. The first frequency spectrum is the representation of the pulse wave bandpass signal in the frequency domain. On the frequency spectrum diagram, the horizontal axis usually represents the frequency, and the vertical axis represents the energy of the signal, showing the energy distribution of the signal at different frequencies. Therefore, by observing the first frequency spectrum, we can intuitively understand which frequency components in the signal are dominant, and which frequency components are weak or almost non-existent. Among them, the peak value usually corresponds to the heart rate value of the user, and at this time the frequency corresponding to the peak value with higher energy in the first frequency spectrum is converted into the second static heart rate value. These first static heart rate values obtained through time domain analysis and second static heart rate values obtained through frequency domain analysis are the static heart rate values of the user in the static state.

[0092] S608, when the user is in a motion state, a first frequency spectrum corresponding to the pulse wave bandpass signal is obtained, and a heart rate value search is performed on the first frequency spectrum to obtain a plurality of first motion heart rate values.

[0093] Optionally, when the user is in a motion state, a plurality of first motion heart rate values can also be determined according to the frequency domain analysis method. For step S608, please refer to the detailed description in step S606, which will not be repeated here.

[0094] S610, and based on the pulse wave signal and the acceleration signal, a deep learning analysis is performed to obtain a second motion heart rate value.

[0095] Optionally, for step S610, please refer to the detailed description in step S206, which will not be repeated here.

[0096] S612, determining the target heart rate value of the user according to each first static heart rate value and each second static heart rate value, or determining the target heart rate value of the user according to each first motion heart rate value and the second motion heart rate value.

[0097] Optionally, for step S612, please refer to the detailed description in step S208, which will not be repeated here.

[0098] In the embodiment of the present application, a heart rate detection method is provided. By simultaneously measuring the acceleration in three directions through a three-axis acceleration sensor, the error caused by single direction measurement can be reduced, thereby providing more accurate acceleration information and improving the overall detection accuracy. By extracting multiple activity parameters in the acceleration band communication signal and comprehensively considering their changes, the current state of the user can be more accurately determined, further improving the accuracy of heart rate detection. By performing fast Fourier transform on the pulse wave band communication signal to obtain a first frequency spectrum, the energy distribution of the pulse wave band communication signal at different frequencies can be intuitively displayed, and spectrum analysis is performed accordingly to obtain a second resting heart rate value or a first exercise heart rate value, which helps to reduce the interference of noise and other non-heart rate related signals, thereby improving the accuracy of heart rate detection.

[0099] Please refer to Figure 7 , Figure 7 The flowchart of a heart rate detection method provided in the embodiment of the present application is shown.

[0100] As Figure 7 shown, the heart rate detection method can at least include:

[0101] S702, obtaining a pulse wave signal and an acceleration signal of a user, performing time domain band pass filtering on the pulse wave signal according to a preset heart rate frequency band to obtain a pulse wave band communication signal, performing time domain band pass filtering on the acceleration signal according to the preset heart rate frequency band to obtain an acceleration band communication signal, determining activity parameters of the user according to the acceleration band communication signal, determining that the user is in a resting state when the activity parameters are less than a preset threshold, and determining that the user is in an exercise state when the activity parameters are greater than or equal to the preset threshold.

[0102] Optionally, for step S702, please refer to the detailed description in step S602, which will not be repeated here.

[0103] S704, when the user is in a resting state, performing time domain analysis based on the pulse wave signal to obtain at least one first resting heart rate value.

[0104] Optionally, for step S704, please refer to the detailed description in step S204, which will not be repeated here.

[0105] S706, obtaining the pulse wave power spectral density corresponding to the pulse wave band communication signal, obtaining the acceleration power spectral density corresponding to the acceleration band communication signal, and obtaining the prior signal-to-noise ratio and the posterior signal-to-noise ratio of the acceleration band communication signal corresponding to the pulse wave band communication signal according to the pulse wave power spectral density and the acceleration power spectral density.

[0106] Optionally, acceleration signals can reflect the user's exercise intensity and movement pattern, information crucial for accurate heart rate calculation. For example, in different types of exercise such as running, walking, or cycling, pulse wave signals may be subject to varying degrees of interference, while acceleration signals can provide useful information about the type and intensity of exercise, thus helping to estimate heart rate more accurately. Therefore, the acquired acceleration signal can not only be used to determine the user's current state but also in frequency domain analysis methods (applicable to frequency domain analysis methods for motion states), combining it with pulse wave signals to improve the accuracy of heart rate detection. Based on this, Figure 8 This is a flowchart illustrating the frequency domain analysis performed in a heart rate detection method according to an embodiment of this application. Figure 8 As shown in S802, in the frequency domain analysis method under static conditions, the pulse wave bandpass signal and the multi-channel acceleration bandpass signal (x, y, z acceleration bandpass signals) are first obtained.

[0107] Optionally, such as Figure 8 As shown in S804, in frequency domain analysis methods, to quantify the energy distribution of pulse wave bandpass signals and acceleration bandpass signals within a specific frequency range, and to assess the noise level in the signals, power spectral density (PSD) can be established for these two signals separately. Power spectral density is a physical quantity describing the change of signal power with frequency; it reflects the energy distribution of the signal at different frequencies. Since acceleration bandpass signals contain motion-related noise, by obtaining the multi-channel acceleration power spectral density of multiple acceleration bandpass signals, we can assess in which frequency ranges this noise is most significant. Therefore, by comparing the pulse wave power spectral density and the acceleration power spectral density, we can more accurately identify the frequency components related to heartbeats and avoid misinterpreting motion noise as heart rate signals. Simultaneously, based on this information, suitable filters can be designed to remove noise components while retaining useful heart rate information, improving the robustness and accuracy of heart rate detection.

[0108] Optionally, such as Figure 8 As shown in S806, the prior signal-to-noise ratio (Prior SNR) of the multiple acceleration bandpass signals corresponding to the pulse wave bandpass signal is first obtained from the power spectral density. This SNR reflects the strength of the pulse wave bandpass signal relative to the acceleration noise. Then, the signal is filtered by Wiener filtering in the frequency domain analysis, and the corresponding posterior signal-to-noise ratio (Posterior SNR) is obtained again. This allows for subsequent filtering of the first spectrum of the pulse wave bandpass signal based on the prior SNR and in combination with the posterior SNR.

[0109] It should be noted that when the acceleration band communication signal has three paths of x, y, and z, each frequency point of each pulse wave band communication signal corresponds to the signal-to-noise ratio of the three paths of x, y, and z, respectively, to realize the optimization of the frequency spectrum through the multi-filter cascade form. For example, assuming that 100 frequency points are set in the pulse wave band communication signal of 1-2 Hz (indicated as a frequency point matrix 【1, 100】, 1.01 Hz, 1.02 Hz, 1.03 Hz, etc.), each frequency point corresponds to the signal-to-noise ratio of the three paths of x, y, and z (indicated as an x-path signal-to-noise ratio matrix 【1, 100】, a y-path signal-to-noise ratio matrix 【1, 100】, and a z-path signal-to-noise ratio matrix 【1, 100】), and the pulse wave band communication signal at each frequency point can be filtered in real time according to the signal-to-noise ratio at each frequency point. Therefore, the number of pulse wave band communication signals (such as one path, two paths, etc.) and the frequency point resolution (such as 10 frequency points set in 1 Hz, with a resolution of 0.1 Hz; 100 frequency points, with a resolution of 0.01 Hz) can be set according to actual needs. For example, on a wearable device such as a sports watch, considering the problem of saving power consumption, only one pulse wave band communication signal can be obtained, and a lower resolution frequency point can be set.

[0110] S708, obtaining a first frequency spectrum corresponding to the pulse wave band communication signal; filtering the first frequency spectrum according to the prior signal-to-noise ratio and the posterior signal-to-noise ratio to obtain a second frequency spectrum corresponding to the pulse wave band communication signal, and searching for a second static heart rate value from the second frequency spectrum.

[0111] Optionally, according to the power spectral density and the signal-to-noise ratio information, the first frequency spectrum corresponding to the pulse wave band communication signal can be filtered. Specifically, the filtering process can perform weighted processing on different frequency components in the frequency spectrum according to the signal-to-noise ratio. In the frequency band with a high signal-to-noise ratio (usually the range corresponding to the heart rate), more signal components are retained; in the frequency band with a low signal-to-noise ratio (usually the range corresponding to the noise), more noise components are suppressed. In this way, the frequency spectrum can be further optimized, and the accuracy of heart rate detection can be improved. After the frequency spectrum filtering based on the signal-to-noise ratio, a second frequency spectrum of the pulse wave band communication signal is obtained. This frequency spectrum is more pure than the first frequency spectrum, and the noise components are effectively suppressed. Next, according to the frequency component and intensity distribution in the second frequency spectrum, a possible second static heart rate value is calculated using a heart rate detection algorithm.

[0112] S710, when the user is in a motion state, obtaining a pulse wave power spectral density corresponding to the pulse wave band communication signal and an acceleration power spectral density corresponding to the acceleration band communication signal; obtaining a prior signal-to-noise ratio and a posterior signal-to-noise ratio of the acceleration band communication signal corresponding to the pulse wave band communication signal according to the pulse wave power spectral density and the acceleration power spectral density.

[0113] Optionally, for step S710, please refer to the detailed description in step S706, which will not be repeated here.

[0114] S712, obtain a first spectrum corresponding to the pulse wave band communication signal; filter the first spectrum according to the prior SNR and the posterior SNR to obtain a second spectrum corresponding to the pulse wave band communication signal, and perform heart rate value searching on the second spectrum to obtain a plurality of first exercise heart rate values.

[0115] Optionally, the first spectrum in the exercise state is also filtered according to the calculation of the SNR to obtain a second spectrum that is more pure, and then the related first exercise heart rate value is obtained from the second spectrum. Specifically, for step S712, please refer to the detailed description in step S708, which will not be repeated here.

[0116] S714, and obtain an acceleration spectrum corresponding to the acceleration band communication signal; input the second spectrum, the acceleration spectrum and the activity parameter into a preset deep learning model; obtain an exercise heart rate value output by the deep learning model to obtain a second exercise heart rate value.

[0117] Optionally, Figure 9 A flowchart of a deep learning analysis process in a heart rate detection method provided by an embodiment of the present application is shown in Figure 9 In the deep learning analysis method in the exercise state, in order to convert the acceleration band communication signal from the time domain to the frequency domain so as to combine and analyze it with the pulse wave band communication signal in the frequency domain, first, the acceleration band communication signal is subjected to spectrum analysis to obtain an acceleration spectrum. Then, as shown in S904-S906 in Figure 9 The second spectrum, the acceleration spectrum and the activity parameter are input as parameters into the trained deep learning model, and the second exercise heart rate value is obtained through the model. Specifically, for how to perform deep learning analysis, please refer to the detailed description in step S206, which will not be repeated here.

[0118] Optionally, the first exercise heart rate value obtained through the frequency domain analysis and the second exercise heart rate value obtained through the deep learning analysis are taken as the exercise heart rate value in the exercise state.

[0119] S716, determine the target heart rate value of the user according to the first resting heart rate values and the second resting heart rate values, or determine the target heart rate value of the user according to the first exercise heart rate values and the second exercise heart rate values.

[0120] Optionally, for step S716, please refer to the detailed description in step S208, which will not be repeated here.

[0121] In the embodiment of the present application, a heart rate detection method is provided, the priori signal-to-noise ratio and the posteriori signal-to-noise ratio are calculated by using the pulse wave power spectral density and the acceleration power spectral density, and the first spectrum of the pulse wave band pass signal is filtered based on the priori signal-to-noise ratio and the posteriori signal-to-noise ratio to obtain the second spectrum, which further optimizes the original spectrum, filters out the frequency components with low signal-to-noise ratio, and retains the frequency components more relevant to the heart rate information, thereby improving the accuracy of heart rate detection; through the deep learning analysis method, the feature extraction and pattern recognition capabilities of the deep learning model can be fully utilized, the features closely related to the heart rate are extracted from the complex signals based on the pulse wave signal, the acceleration signal and the activity parameter, and the heart rate of the user in different states is more accurately identified, thereby improving the accuracy of heart rate detection.

[0122] Please refer to Figure 10 , Figure 10 The flowchart of the heart rate detection method provided in the embodiment of the present application is shown.

[0123] As shown in Figure 10 , the heart rate detection method can at least include:

[0124] S1002, the pulse wave signal and the acceleration signal of the user are acquired, the pulse wave signal is time domain band pass filtered according to the preset heart rate frequency band to obtain the pulse wave band pass signal, the acceleration signal is time domain band pass filtered according to the preset heart rate frequency band to obtain the acceleration band pass signal, the activity parameter of the user is determined according to the acceleration band pass signal, when the activity parameter is less than the preset threshold, it is judged that the user is in a static state, and when the activity parameter is greater than or equal to the preset threshold, it is judged that the user is in a motion state.

[0125] Optionally, for step S1002, please refer to the detailed description in step S602, which will not be repeated here.

[0126] S1004, when the user is in a static state, time domain analysis is performed based on the pulse wave signal to obtain at least one first static heart rate value.

[0127] Optionally, for step S1004, please refer to the detailed description in step S204, which will not be repeated here.

[0128] S1006, the pulse wave power spectral density corresponding to the pulse wave band pass signal is acquired, the acceleration power spectral density corresponding to the acceleration band pass signal is acquired, and the priori signal-to-noise ratio and the posteriori signal-to-noise ratio of the pulse wave band pass signal corresponding to the acceleration band pass signal are obtained according to the pulse wave power spectral density and the acceleration power spectral density.

[0129] Optionally, for step S1006, please refer to the detailed description in step S706, which will not be repeated here.

[0130] S1008. Obtain the first spectrum corresponding to the pulse wave bandpass signal; determine the first frequency domain filtering coefficients through the prior signal-to-noise ratio and the posterior signal-to-noise ratio, and filter the first spectrum through the first frequency domain filtering coefficients; or determine the first frequency domain filtering coefficients through the prior signal-to-noise ratio and the posterior signal-to-noise ratio, determine the second frequency domain filtering coefficients through the first frequency domain filtering coefficients and the activity parameters, filter the first spectrum according to the second frequency domain filtering coefficients, and obtain the second spectrum corresponding to the pulse wave bandpass signal.

[0131] Optionally, Figure 8 This is a flowchart illustrating the frequency domain analysis performed in a heart rate detection method according to an embodiment of this application. Figure 8 As shown in S808, in one feasible implementation, when optimizing the first spectrum, algorithms such as Wiener filtering can be used to construct and dynamically adjust multiple first frequency domain filter coefficients based on the changes in the prior and posterior signal-to-noise ratios. Each pulse wave bandpass signal corresponds to filter coefficients on the x, y, and z acceleration bandpass signals, respectively. For example, the x-channel first frequency domain filter coefficient matrix [1, n] (where n is the number of frequency points representing the resolution). Then, the calculated multiple first frequency domain filter coefficients are used to filter the first spectrum (the multiple first frequency domain filter coefficient matrix [1, n] is multiplied point by point with the frequency point matrix [1, n]) to obtain the optimized second spectrum, thereby effectively suppressing noise while preserving the heart rate-related signal.

[0132] Optionally, such as Figure 8 As shown in S810-814, in another feasible implementation, the user's activity parameters are further introduced to adjust the second spectrum based on the multi-channel first frequency domain filtering coefficients. Specifically, the multi-channel first frequency domain filtering coefficients are further optimized in conjunction with the user's activity parameters to obtain the corresponding multi-channel second frequency domain filtering coefficients. This optimization process can be an adjustment based on preset rules, or it can be a pre-trained model that can automatically adjust the filtering coefficients according to different activity parameters to adapt to the heart rate detection requirements under different exercise states. Then, the multi-channel second frequency domain filtering coefficients are applied to filter the first spectrum of the pulse wave bandpass signal to obtain the optimized second spectrum.

[0133] S1010. Based on the activity parameters, perform peak search on the second spectrum to determine multiple heart rate values ​​to be confirmed; correct each heart rate value to be confirmed using a resolution refinement method to obtain multiple second resting heart rate values.

[0134] Optionally, in frequency domain analysis, the second spectrum is represented with frequency on the horizontal axis and energy on the vertical axis, showing the energy distribution of the signal at different frequencies. High-energy frequencies are most likely to correspond to the true heart rate, while noise typically manifests as a widely distributed low-energy component. Since we are interested in the spectral components corresponding to the heart rate, therefore... Figure 8 As shown in S816, high-energy frequency points related to heart rate can be identified through peak search. However, since various noise interferences may not be completely filtered out during the aforementioned filtering process, these noises may cause multiple high-energy peaks to appear on the second spectrum, not just the peak corresponding to the heart rate. Therefore, simply finding the largest peak may not accurately reflect the true heart rate value. Multiple possible peak positions (i.e., the heart rate value to be confirmed) can be identified to increase the probability of identifying the correct heart rate value. Furthermore, when performing peak search on the second spectrum, activity parameters can be combined to narrow the search range and improve the efficiency of peak search. For example, when the known motion parameters are large, the heart rate usually increases accordingly. If the heart rate value corresponding to a certain peak position is significantly low, then the peak is likely caused by noise rather than the true heart rate value. In this case, a higher search threshold can be set based on the activity parameters to avoid unnecessary searches in the low-frequency region, further improving the accuracy of heart rate detection.

[0135] Optionally, in practical applications, due to limitations such as power consumption considerations, the spectrum may not be subdivided into many frequency points. This means that within a limited frequency range, it may be impossible to accurately capture the true frequency corresponding to the heart rate value. For example, if there are only 10 frequency points in the 1-2 Hz range, with a resolution of 0.1 Hz, then within this range we can only identify specific heart rate values ​​such as 66, 72, and 78 (beats / minute). Obviously, this resolution is insufficient for accurate heart rate measurement, because the true heart rate value may lie between these discrete points. Based on this, as... Figure 8 As shown in S818-S820, to overcome the limitations of the second spectral resolution, we need to refine the resolution of the heart rate value to be confirmed, thereby determining the heart rate value more accurately. Specifically, this can be achieved through various techniques, including at least one of Zoom-FFT transform, Chirp-Z transform, and Yip-Zoom transform. After refining the resolution of the second spectrum, the position of the heart rate value to be confirmed is corrected based on the refinement result, resulting in a corrected second resting heart rate value, which is closer to the user's true heart rate. The first resting heart rate value obtained through time-domain analysis and the second resting heart rate value obtained through frequency-domain analysis are the user's resting heart rate values ​​in a resting state.

[0136] S1012, when the user is in a motion state, acquiring a pulse wave band communication signal corresponding to the pulse wave power spectrum density, acquiring an acceleration band communication signal corresponding to the acceleration power spectrum density; according to the pulse wave power spectrum density and the acceleration power spectrum density, acquiring the prior signal-to-noise ratio and the posterior signal-to-noise ratio of the acceleration band communication signal corresponding to the pulse wave band communication signal.

[0137] Optionally, for step S1012, please refer to the detailed description in step S706, which will not be repeated here.

[0138] S1014, acquiring a first frequency spectrum corresponding to the pulse wave band communication signal; determining a first frequency domain filtering coefficient through the prior signal-to-noise ratio and the posterior signal-to-noise ratio, and filtering the first frequency spectrum through the first frequency domain filtering coefficient; or determining a first frequency domain filtering coefficient through the prior signal-to-noise ratio and the posterior signal-to-noise ratio, and determining a second frequency domain filtering coefficient through the first frequency domain filtering coefficient and the activity parameter, filtering the first frequency spectrum according to the second frequency domain filtering coefficient to obtain a second frequency spectrum corresponding to the pulse wave band communication signal; performing peak search on the second frequency spectrum according to the activity parameter to determine a plurality of to-be-confirmed heart rate values; correcting each to-be-confirmed heart rate value through a resolution refinement method to obtain a plurality of first motion heart rate values; and performing deep learning analysis based on the pulse wave signal and the acceleration signal to obtain a second motion heart rate value.

[0139] Optionally, in the motion state, a plurality of first motion heart rate values can also be obtained according to frequency domain analysis. Specifically, for step S1014, please refer to the detailed description in steps S1008, S1010 and S206, which will not be repeated here.

[0140] S1016, determining a target heart rate value of the user according to each first static heart rate value and each second static heart rate value, or determining a target heart rate value of the user according to each first motion heart rate value and the second motion heart rate value.

[0141] Optionally, for step S1016, please refer to the detailed description in step S208, which will not be repeated here.

[0142] In this embodiment, a heart rate detection method is provided. A first frequency domain filtering coefficient is determined based on prior and posterior signal-to-noise ratios, which can more effectively identify and suppress interference signals generated during exercise, thereby improving the accuracy of heart rate detection. Based on the first frequency domain filtering coefficient, a second frequency domain filtering coefficient is obtained by combining activity parameters. This allows for more refined filtering of the first spectrum for signal characteristics under different exercise states, further optimizing the filtering effect. Next, peak searching is performed on the second spectrum using activity parameters to determine the heart rate value to be confirmed, which can initially screen out heart rate-related signals, laying the foundation for subsequent heart rate value determination. Then, a resolution refinement method is used to correct each heart rate value to obtain the final heart rate value, further improving the accuracy of the heart rate value and reducing errors.

[0143] Please see Figure 11 , Figure 11 This is a structural block diagram of a heart rate detection device provided in an embodiment of this application. Figure 11 As shown, the heart rate detection device 1100 includes:

[0144] The state determination unit 1110 is used to acquire the user's pulse wave signal and acceleration signal, and determine the user's current state, which includes a static state or a moving state.

[0145] The first calculation unit 1120 is used to perform time-domain analysis based on the pulse wave signal to obtain at least one first resting heart rate value when the user is in a resting state, and to perform frequency-domain analysis based on the pulse wave signal to obtain multiple second resting heart rate values.

[0146] The second calculation unit 1130 is used to perform frequency domain analysis based on the pulse wave signal to obtain multiple first exercise heart rate values ​​when the user is in motion, and to perform deep learning analysis based on the pulse wave signal and acceleration signal to obtain second exercise heart rate values.

[0147] The result determination unit 1140 is used to determine the user's target heart rate value based on each first resting heart rate value and each second resting heart rate value, or to determine the user's target heart rate value based on each first exercise heart rate value and each second exercise heart rate value.

[0148] Optionally, the heart rate detection device 1100 further includes: a bandpass filtering unit, used to perform time-domain bandpass filtering on the pulse wave signal according to a preset heart rate frequency band to obtain a pulse wave bandpass signal; and a first calculation unit 1120, used to perform peak and trough search and / or establish a probability density function on the pulse wave bandpass signal to obtain at least one first resting heart rate value.

[0149] Optionally, the heart rate detection apparatus 1100 further comprises a signal difference unit configured to, after the band-pass filtering unit obtains the pulse wave band-pass signal, perform a difference operation on the pulse wave band-pass signal to obtain a pulse wave difference signal; and perform peak-valley searching and / or establishing a probability density function on the pulse wave difference signal to obtain at least one first resting heart rate value.

[0150] Optionally, the first calculation unit 1120 or the second calculation unit 1130 is further configured to obtain a first frequency spectrum corresponding to the pulse wave band-pass signal, and perform heart rate value searching on the first frequency spectrum.

[0151] Optionally, the band-pass filtering unit is further configured to perform time domain band-pass filtering on the acceleration signal according to a preset heart rate frequency band to obtain an acceleration band-pass signal; and the first calculation unit 1120 or the second calculation unit 1130 is further configured to obtain a pulse wave power spectral density corresponding to the pulse wave band-pass signal, obtain an acceleration power spectral density corresponding to the acceleration band-pass signal, obtain a prior signal-to-noise ratio and a posterior signal-to-noise ratio of the acceleration band-pass signal corresponding to the pulse wave band-pass signal according to the pulse wave power spectral density and the acceleration power spectral density, and perform filtering on the first frequency spectrum according to the prior signal-to-noise ratio and the posterior signal-to-noise ratio to obtain a second frequency spectrum corresponding to the pulse wave band-pass signal, and perform heart rate value searching on the second frequency spectrum.

[0152] Optionally, the first calculation unit 1120 or the second calculation unit 1130 is further configured to determine a first frequency domain filtering coefficient through the prior signal-to-noise ratio and the posterior signal-to-noise ratio, and perform filtering on the first frequency spectrum through the first frequency domain filtering coefficient.

[0153] Optionally, the heart rate detection apparatus 1100 further comprises an activity parameter determination unit configured to, after the band-pass filtering unit performs time domain band-pass filtering on the acceleration signal according to a preset heart rate frequency band to obtain an acceleration band-pass signal, determine an activity parameter of a user according to the acceleration band-pass signal; and the first calculation unit 1120 or the second calculation unit 1130 is further configured to determine a first frequency domain filtering coefficient through the prior signal-to-noise ratio and the posterior signal-to-noise ratio, determine a second frequency domain filtering coefficient through the first frequency domain filtering coefficient and the activity parameter, and perform filtering on the first frequency spectrum according to the second frequency domain filtering coefficient.

[0154] Optionally, the first calculation unit 1120 or the second calculation unit 1130 is further configured to perform peak searching on the second frequency spectrum according to the activity parameter to determine a plurality of to-be-confirmed heart rate values, and correct each to-be-confirmed heart rate value through a resolution refinement method.

[0155] Optionally, the second calculation unit 1130 is further configured to obtain an acceleration frequency spectrum corresponding to the acceleration band-pass signal, input the second frequency spectrum, the acceleration frequency spectrum, and the activity parameter into a preset deep learning model, and obtain a motion heart rate value output by the deep learning model.

[0156] Optionally, the state determining unit 1110 is further configured to determine that the user is in a static state when the activity parameter is less than a preset threshold, and determine that the user is in a dynamic state when the activity parameter is greater than or equal to the preset threshold.

[0157] In the embodiments of the present application, a heart rate detection device is provided, wherein a state determining unit is configured to acquire a pulse wave signal and an acceleration signal of a user, and determine a current state of the user, the current state including a static state or a dynamic state; a first calculation unit is configured to, when the user is in the static state, perform time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, and perform frequency domain analysis based on the pulse wave signal to obtain a plurality of second static heart rate values; a second calculation unit is configured to, when the user is in the dynamic state, perform frequency domain analysis based on the pulse wave signal to obtain a plurality of first dynamic heart rate values, and perform deep learning analysis based on the pulse wave signal and the acceleration signal to obtain a second dynamic heart rate value; and a result determining unit is configured to determine a target heart rate value of the user according to the first static heart rate values and the second static heart rate values, or determine the target heart rate value of the user according to the first dynamic heart rate values and the second dynamic heart rate value. When performing heart rate detection, the pulse wave signal and the acceleration signal are first collected by the state determining unit, which provides a data basis for subsequent state determination and heart rate analysis. The acceleration signal can effectively assist in distinguishing between pulse wave changes caused by movement and real heart rate changes, which helps to more comprehensively understand the heart rate changes of the user. At the same time, accurately determining the current state of the user is a prerequisite for selecting an appropriate heart rate detection method subsequently, which can ensure that time domain analysis and frequency domain analysis that reduce power consumption are used in the static state, while frequency domain analysis and deep learning analysis are more effectively used in the dynamic state to reduce the influence of motion artifacts. When the user is in the static state, the first calculation unit performs time domain analysis based on the pulse wave signal, which can directly reflect the trend of the pulse wave change to obtain basic information of the heart rate, and at the same time, the pulse wave signal is converted from the time domain to the frequency domain to observe the distribution of the pulse wave at different frequencies, thereby obtaining a plurality of heart rate values in the static state. When the user is in the dynamic state, the second calculation unit performs frequency domain analysis and introduces a deep learning model to comprehensively analyze the pulse wave signal and the acceleration signal. The deep learning model can learn and identify the pulse wave change pattern caused by movement, thereby effectively reducing the influence of motion artifacts on heart rate detection and more accurately extracting real heart rate information in the dynamic state. Finally, the result determining unit determines the target heart rate value of the user according to the heart rate values in the static state or the heart rate values in the dynamic state, which can help the user understand his or her heart rate condition and serve as an important basis for health management and disease prevention.

[0158] The embodiments of the present application further provide a computer storage medium, which can store a plurality of instructions, the instructions being suitable for being loaded and executed by a processor to perform the steps of the method of any one of the above embodiments.

[0159] Please refer to Figure 12 , Figure 12 A structural schematic diagram of an electronic device is provided for an embodiment of the present application. As shown in Figure 12 , the electronic device 1200 can include at least one processor 1201, at least one network interface 1204, a user interface 1203, a memory 1205, and at least one communication bus 1202.

[0160] The communication bus 1202 is configured to realize the connection and communication between the components.

[0161] The user interface 1203 can include a display screen (Display), a camera (Camera), and can also include a standard wired interface and a wireless interface.

[0162] The network interface 1204 can optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0163] The processor 1201 can include one or more processing cores. The processor 1201 connects various parts of the entire electronic device 1200 through various interfaces and lines, and performs various functions and processes data of the electronic device 1200 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1205, and calling data stored in the memory 1205. Optionally, the processor 1201 can be implemented in at least one of a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 1201 can be integrated with a combination of one or more of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. Among them, the CPU is mainly used to process the operating system, user interface, and application program; the GPU is used to render and draw the content to be displayed on the display screen; and the modem is used to process wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor 1201, but can be realized by a separate chip.

[0164] The memory 1205 can include a random access memory (RAM) and can also include a read-only memory (ROM). Optionally, the memory 1205 includes a non-transitory computer-readable storage medium. The memory 1205 can be used to store instructions, programs, codes, code sets, or instruction sets. The memory 1205 can include a program storage area and a data storage area, where the program storage area can store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store data involved in the above-mentioned various method embodiments, etc. The memory 1205 can also be at least one storage device located away from the aforementioned processor 1201. As shown in Figure 12 The memory 1205 as a computer storage medium can include an operating system, a network communication module, a user interface module, and a heart rate detection program.

[0165] In the electronic device 1200 shown in Figure 12 In the electronic device 1200 shown in

[0166] Obtaining a pulse wave signal and an acceleration signal of a user, determining a current state of the user, the current state including a static state or a motion state;

[0167] When the user is in the static state, performing time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, and performing frequency domain analysis based on the pulse wave signal to obtain a plurality of second static heart rate values;

[0168] When the user is in the motion state, performing frequency domain analysis based on the pulse wave signal to obtain a plurality of first motion heart rate values, and performing deep learning analysis based on the pulse wave signal and the acceleration signal to obtain a second motion heart rate value;

[0169] Determining a target heart rate value of the user according to the first static heart rate values and the second static heart rate values, or determining the target heart rate value of the user according to the first motion heart rate values and the second motion heart rate value.

[0170] In some embodiments, after obtaining the pulse wave signal and the acceleration signal of the user, the processor 1201 further specifically performs the following steps: performing time domain band-pass filtering on the pulse wave signal according to a preset heart rate frequency band to obtain a pulse wave band-pass signal; when performing time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, the processor 1201 specifically performs the following steps: performing peak-valley searching on the pulse wave band-pass signal and / or establishing a probability density function to obtain the at least one first static heart rate value.

[0171] In some embodiments, after obtaining the pulse wave signal and the acceleration signal of the user, the processor 1201 further specifically performs the following steps: performing time domain band-pass filtering on the pulse wave signal according to a preset heart rate frequency band to obtain a pulse wave band-pass signal; when performing time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, the processor 1201 specifically performs the following steps: performing peak-valley searching on the pulse wave band-pass signal and / or establishing a probability density function to obtain the at least one first static heart rate value.

[0172] In some embodiments, after obtaining the pulse wave signal and the acceleration signal of the user, the processor 1201 further specifically performs the following steps: performing time domain band-pass filtering on the pulse wave signal according to a preset heart rate frequency band to obtain a pulse wave band-pass signal; when performing time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, the processor 1201 specifically performs the following steps: performing peak-valley searching on the pulse wave band-pass signal and / or establishing a probability density function to obtain the at least one first static heart rate value.

[0173] In some embodiments, after obtaining the pulse wave signal and the acceleration signal of the user, the processor 1201 further specifically performs the following steps: performing time domain band-pass filtering on the pulse wave signal according to a preset heart rate frequency band to obtain a pulse wave band-pass signal; when performing time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, the processor 1201 specifically performs the following steps: performing peak-valley searching on the pulse wave band-pass signal and / or establishing a probability density function to obtain the at least one first static heart rate value.

[0174] According to the pulse wave power spectral density and the acceleration power spectral density, obtaining a prior signal-to-noise ratio and a posterior signal-to-noise ratio of the pulse wave band-pass signal corresponding to the acceleration band-pass signal;

[0175] According to the prior signal-to-noise ratio and the posterior signal-to-noise ratio, filtering the first frequency spectrum to obtain a second frequency spectrum corresponding to the pulse wave band-pass signal, and performing heart rate value searching on the second frequency spectrum.

[0176] In some embodiments, when performing filtering on the first frequency spectrum according to the prior signal-to-noise ratio and the posterior signal-to-noise ratio, the processor 1201 specifically performs the following steps: determining a first frequency domain filtering coefficient through the prior signal-to-noise ratio and the posterior signal-to-noise ratio, and filtering the first frequency spectrum through the first frequency domain filtering coefficient.

[0177] In some embodiments, after performing the time domain band-pass filtering on the acceleration signal according to the preset heart rate frequency band to obtain the acceleration band-pass signal, the processor 1201 further specifically performs the following steps: determining the activity parameter of the user according to the acceleration band-pass signal; when performing the filtering on the first frequency spectrum according to the prior signal-to-noise ratio and the posterior signal-to-noise ratio, the processor 1201 specifically performs the following steps: determining the first frequency domain filtering coefficient through the prior signal-to-noise ratio and the posterior signal-to-noise ratio, and determining the second frequency domain filtering coefficient through the first frequency domain filtering coefficient and the activity parameter, and filtering the first frequency spectrum according to the second frequency domain filtering coefficient.

[0178] In some embodiments, when performing the heart rate value lookup on the second frequency spectrum, the processor 1201 specifically performs the following steps: performing a peak search on the second frequency spectrum according to the activity parameter to determine a plurality of to-be-confirmed heart rate values; and correcting each to-be-confirmed heart rate value through a resolution refinement method.

[0179] In some embodiments, when performing the deep learning analysis based on the pulse wave signal and the acceleration signal, the processor 1201 specifically performs the following steps: obtaining an acceleration frequency spectrum corresponding to the acceleration band-pass signal; inputting the second frequency spectrum, the acceleration frequency spectrum, and the activity parameter into a preset deep learning model; and obtaining a motion heart rate value output by the deep learning model.

[0180] In some embodiments, when performing the judgment on the current state of the user, the processor 1201 specifically performs the following steps: when the activity parameter is less than a preset threshold, judging that the user is in a static state; and when the activity parameter is greater than or equal to the preset threshold, judging that the user is in a motion state.

[0181] In several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented by other manners. For example, the apparatus embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, a plurality of modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed ones can be indirect coupling or communication connection through some interfaces, apparatuses or modules, which can be electrical, mechanical or other forms.

[0182] The modules illustrated as separate components can or can not be physically separate, and the components illustrated as modules can or can not be physical modules, i.e., can be located in one place, or can be distributed on a plurality of network modules. Part or all of the modules can be selected according to actual needs to achieve the purpose of the present embodiment.

[0183] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The above computer program product includes one or more computer instructions. When the above computer program instructions are loaded and executed on a computer, all or part of the processes or functions described above according to the embodiments of the present disclosure are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted by the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available media can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a digital versatile disc (DVD)), or a semiconductor medium (for example, a solid state disk (SSD)) and the like.

[0184] It should be noted that for the foregoing method embodiments, in order to facilitate description, they are all described as a combination of a series of actions, but those skilled in the art should know that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0185] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0186] The above is the description of the heart rate detection method, device, storage medium and electronic equipment provided by the present application. For those skilled in the art, according to the idea of the embodiments of the present application, there will be changes in specific implementation and application range. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A heart rate detection method, characterized by, The method comprises: acquiring a pulse wave signal and an acceleration signal of a user, judging a current state of the user, the current state comprising a static state or a motion state; performing time domain band-pass filtering on the pulse wave signal according to a preset heart rate frequency band to obtain a pulse wave band-pass signal, and performing time domain band-pass filtering on the acceleration signal according to the preset heart rate frequency band to obtain an acceleration band-pass signal; when the user is in the static state, performing time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, acquiring a first frequency spectrum corresponding to the pulse wave band-pass signal, acquiring a pulse wave power spectral density corresponding to the pulse wave band-pass signal, and acquiring an acceleration power spectral density corresponding to the acceleration band-pass signal; acquiring a prior signal-to-noise ratio and a posterior signal-to-noise ratio of the pulse wave band-pass signal corresponding to the acceleration band-pass signal according to the pulse wave power spectral density and the acceleration power spectral density; filtering the first frequency spectrum according to the prior signal-to-noise ratio and the posterior signal-to-noise ratio to obtain a second frequency spectrum corresponding to the pulse wave band-pass signal, and performing heart rate value lookup on the second frequency spectrum to obtain a plurality of second static heart rate values; when the user is in the motion state, acquiring a first frequency spectrum corresponding to the pulse wave band-pass signal, acquiring a pulse wave power spectral density corresponding to the pulse wave band-pass signal, and acquiring an acceleration power spectral density corresponding to the acceleration band-pass signal; acquiring a prior signal-to-noise ratio and a posterior signal-to-noise ratio of the pulse wave band-pass signal corresponding to the acceleration band-pass signal according to the pulse wave power spectral density and the acceleration power spectral density; filtering the first frequency spectrum according to the prior signal-to-noise ratio and the posterior signal-to-noise ratio to obtain a second frequency spectrum corresponding to the pulse wave band-pass signal, and performing heart rate value lookup on the second frequency spectrum to obtain a plurality of first motion heart rate values, and performing deep learning analysis based on the pulse wave signal and the acceleration signal to obtain a second motion heart rate value; determining a target heart rate value of the user according to the first static heart rate values and the second static heart rate values, or determining the target heart rate value of the user according to the first motion heart rate values and the second motion heart rate value.

2. The method of claim 1, wherein, After acquiring the pulse wave signal and the acceleration signal of the user, the method further comprises: performing time domain band-pass filtering on the pulse wave signal according to a preset heart rate frequency band to obtain a pulse wave band-pass signal; the time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value comprises: performing peak-valley searching and / or establishing a probability density function on the pulse wave band-pass signal to obtain at least one first static heart rate value.

3. The method of claim 2, wherein, after obtaining the pulse wave band-pass signal, the method further comprises: performing difference on the pulse wave band-pass signal to obtain a pulse wave difference signal; performing peak-valley searching and / or establishing a probability density function on the pulse wave difference signal to obtain at least one first static heart rate value.

4. The method of claim 1, wherein, the filtering the first frequency spectrum according to the prior signal-to-noise ratio and the posterior signal-to-noise ratio comprises: determine a first frequency domain filter coefficient according to the prior SNR and the posterior SNR, and filter the first frequency spectrum according to the first frequency domain filter coefficient.

5. The method of claim 1, wherein, After the acceleration signal is filtered in the time domain according to the preset heart rate frequency band to obtain an acceleration bandpass signal, the method further comprises: determining an activity parameter of the user according to the acceleration bandpass signal; the filtering of the first frequency spectrum according to the prior SNR and the posterior SNR comprises: determining a first frequency domain filter coefficient according to the prior SNR and the posterior SNR, and determining a second frequency domain filter coefficient according to the first frequency domain filter coefficient and the activity parameter, and filtering the first frequency spectrum according to the second frequency domain filter coefficient.

6. The method of claim 5, wherein, the heart rate value searching of the second frequency spectrum comprises: performing a peak searching on the second frequency spectrum according to the activity parameter to determine a plurality of to-be-confirmed heart rate values; correcting each to-be-confirmed heart rate value by a resolution refinement method.

7. The method of claim 5, wherein, the deep learning analysis based on the pulse wave signal and the acceleration signal comprises: obtaining an acceleration frequency spectrum corresponding to the acceleration bandpass signal; inputting the second frequency spectrum, the acceleration frequency spectrum, and the activity parameter into a preset deep learning model; obtaining a motion heart rate value output by the deep learning model.

8. The method of claim 5, wherein, the determination of the current state of the user comprises: when the activity parameter is less than a preset threshold, determining that the user is in a static state; when the activity parameter is greater than or equal to the preset threshold, determining that the user is in a motion state.

9. A heart rate detection apparatus characterized by comprising: the device comprises: a state determination unit configured to obtain a pulse wave signal and an acceleration signal of a user, and determine a current state of the user, the current state comprising a static state or a motion state; a bandpass filtering unit configured to filter the pulse wave signal in the time domain according to a preset heart rate frequency band to obtain a pulse wave bandpass signal, and filter the acceleration signal in the time domain according to the preset heart rate frequency band to obtain an acceleration bandpass signal; a first calculation unit configured to, when the user is in the static state, perform a time domain analysis based on the pulse wave signal to obtain at least one first static heart rate value, obtain a first frequency spectrum corresponding to the pulse wave bandpass signal, obtain a pulse wave power spectral density corresponding to the pulse wave bandpass signal, and obtain an acceleration power spectral density corresponding to the acceleration bandpass signal; determine a prior SNR and a posterior SNR of the acceleration bandpass signal corresponding to the pulse wave bandpass signal according to the pulse wave power spectral density and the acceleration power spectral density; filter the first frequency spectrum according to the prior SNR and the posterior SNR to obtain a second frequency spectrum corresponding to the pulse wave bandpass signal, and perform a heart rate value searching on the second frequency spectrum to obtain a plurality of second static heart rate values. The second computing unit is configured to, when the user is in a motion state, acquire a first frequency spectrum corresponding to the pulse wave band communication signal, acquire a pulse wave power spectral density corresponding to the pulse wave band communication signal, and acquire an acceleration power spectral density corresponding to the acceleration band communication signal; acquire a prior signal-to-noise ratio and a posterior signal-to-noise ratio of the pulse wave band communication signal corresponding to the acceleration band communication signal according to the pulse wave power spectral density and the acceleration power spectral density; filter the first frequency spectrum according to the prior signal-to-noise ratio and the posterior signal-to-noise ratio to obtain a second frequency spectrum corresponding to the pulse wave band communication signal, perform heart rate value lookup on the second frequency spectrum to obtain a plurality of first motion heart rate values, and perform deep learning analysis based on the pulse wave signal and the acceleration signal to obtain a second motion heart rate value. The result determining unit is configured to determine a target heart rate value of the user according to the first static heart rate values and the second static heart rate values, or determine the target heart rate value of the user according to the first motion heart rate values and the second motion heart rate value.

10. A computer storage medium, characterized in that, The computer storage medium stores a plurality of instructions, which are suitable for being loaded and executed by the processor to perform the steps of the method according to any one of claims 1-8.

11. An electronic device, comprising: The computer program is stored in the memory and executable on the processor, and the processor executes the program to implement the steps of the method according to any one of claims 1-8.

Images