Frequency determination method, device, non-volatile storage medium and computer equipment

Abstract

The present invention discloses a frequency determination method, apparatus, non-volatile storage medium, and computer equipment. The method comprises: obtaining a first frequency value of a target object, wherein the first frequency value represents the frequency of the target object's vital signs at the end of a first time window; obtaining an optical power sequence signal, wherein the optical power sequence signal is obtained by an optical fiber sensor monitoring the target object within a second time window, and a first timing relationship between the first time window and the second time window is that the second time window is later than the first time window; inputting the first frequency value, the optical power sequence signal, and the first timing relationship into a frequency determination model to obtain a second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the target object's vital signs at the end of the second time window. The present invention solves the technical problem of difficulty in obtaining accurate monitoring results when using an optical fiber sensor to monitor the frequency of a user's vital signs.

Description

Technical Field

[0001] The present invention relates to the field of signal processing technology, and in particular to a frequency determination method, device, non-volatile storage medium, and computer equipment. Background Art

[0002] Vital sign sensing technologies are mainly divided into skin-contact and non-skin-contact vital sign sensing technologies. In situations where long-term monitoring is required (such as sleep monitoring), skin-contact vital sign sensing technologies can cause users to feel uncomfortable. Some technologies require the use of specific equipment and professional operation, and may be subject to environmental and usage conditions. Compared to skin-contact vital sign sensing technologies, non-skin-contact vital sign sensing technologies do not require direct contact with the body, reducing user discomfort, improving comfort, and providing a more convenient and easier-to-use monitoring method suitable for long-term monitoring and autonomous health management. In addition, some non-contact technologies can provide real-time monitoring data.

[0003] However, some existing non-contact vital sign monitoring technologies lack accuracy. Fiber optic sensors, with their high sensitivity, resistance to electromagnetic interference, small size, safety, and versatility, are widely used in industry, healthcare, communications, and other fields. Fiber optic sensor-based vital sign monitoring is a relatively new type of non-skin contact vital sign monitoring technology. However, due to its high sensitivity, the predicted vital sign frequency can differ significantly from the actual result under the interference of human body movement, posing a significant challenge to fiber optic sensor-based vital sign monitoring technology.

[0004] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0005] Embodiments of the present invention provide a frequency determination method, apparatus, non-volatile storage medium, and computer equipment to at least solve the technical problem of difficulty in obtaining accurate monitoring results when using optical fiber sensors to monitor the frequency of a user's vital signs.

[0006] According to one aspect of an embodiment of the present invention, a frequency determination method is provided, comprising: obtaining a first frequency value of a target object, wherein the first frequency value represents the frequency of a vital sign of the target object at the end of a first time window; obtaining an optical power sequence signal, wherein the optical power sequence signal is obtained by an optical fiber sensor monitoring the target object within a second time window, and a first timing relationship between the first time window and the second time window is that the second time window is later than the first time window; inputting the first frequency value, the optical power sequence signal, and the first timing relationship into a frequency determination model to obtain a second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the vital sign of the target object at the end of the second time window.

[0007] Optionally, inputting the first frequency value, the optical power sequence signal and the first timing relationship into a frequency determination model to obtain a second frequency value output by the frequency determination model includes: inputting the optical power sequence signal into a feature extraction layer in the frequency determination model to obtain a frequency feature output by the feature extraction layer; inputting the first timing relationship and the frequency feature into a splicing layer in the frequency determination model to obtain a splicing vector output by the splicing layer, wherein the splicing vector is obtained by splicing the frequency feature and the first frequency value according to the first timing relationship; and inputting the splicing vector into a prediction layer in the frequency determination model to obtain the second frequency value output by the prediction layer.

[0008] Optionally, the frequency determination model is a model generated in advance through the following steps: obtaining training samples, wherein the training samples include: a first label representing the vital sign frequency of the sample object at a first moment, a second label representing the vital sign frequency of the sample object at a second moment, and an optical power sequence signal sample; the first moment is the end moment of a first sample time window, the second moment is the end moment of a second sample time window, and the optical power sequence signal sample is obtained by the optical fiber sensor monitoring the sample object within the second sample time window; the first label and the optical power sequence signal sample are input into the original model corresponding to the frequency determination model to obtain a vital sign frequency prediction value output by the original model; according to the difference between the vital sign frequency prediction value and the second label, the model parameters of the original model are updated to obtain the frequency determination model.

[0009] Optionally, before obtaining the optical power sequence signal, the method further includes: obtaining a vital sign update cycle; and according to the vital sign update cycle, shifting the first time window backward to obtain the second time window, wherein the time difference between the second time window and the first time window matches the vital sign update cycle.

[0010] Optionally, obtaining the first frequency value of the target object includes: obtaining a prior optical power sequence signal, wherein the prior optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within the first time window; performing a fast Fourier transform on the prior optical power sequence signal to obtain a prior frequency domain signal; and determining the frequency corresponding to the maximum amplitude in the prior frequency domain signal as the first frequency value.

[0011] Optionally, after obtaining the second frequency value output by the frequency determination model, the method further includes: acquiring a subsequent optical power sequence signal, wherein the subsequent optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a third time window, and the second timing relationship between the second time window and the third time window is that the third time window is later than the second time window; inputting the second frequency value, the subsequent optical power sequence signal and the second timing relationship into the frequency determination model to obtain a third frequency value output by the frequency determination model, wherein the third frequency value represents the frequency of the vital signs of the target object at the end moment of the third time window.

[0012] Optionally, the vital sign is breathing or heartbeat.

[0013] According to another aspect of an embodiment of the present invention, a frequency determination device is also provided, including: a first acquisition module, used to acquire a first frequency value of a target object, wherein the first frequency value represents the frequency of the vital sign of the target object at the end of a first time window; a second acquisition module, used to acquire an optical power sequence signal, wherein the optical power sequence signal is obtained by monitoring the target object within a second time window by an optical fiber sensor, and a first timing relationship between the first time window and the second time window is that the second time window is later than the first time window; a determination module, used to input the first frequency value, the optical power sequence signal and the first timing relationship into a frequency determination model to obtain a second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the vital sign of the target object at the end of the second time window.

[0014] According to another aspect of an embodiment of the present invention, a non-volatile storage medium is provided, wherein the non-volatile storage medium includes a stored program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute any one of the above-mentioned frequency determination methods.

[0015] According to another aspect of an embodiment of the present invention, a computer device is provided, comprising a memory and a processor, wherein the memory is used to store programs, and the processor is used to run the programs stored in the memory, wherein the program executes any one of the above-mentioned frequency determination methods when running.

[0016] In an embodiment of the present invention, a first frequency value of a target object is obtained, wherein the first frequency value represents the frequency of the vital sign of the target object at the end moment of a first time window; an optical power sequence signal is obtained, wherein the optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a second time window, and a first timing relationship between the first time window and the second time window is that the second time window is later than the first time window; the first frequency value, the optical power sequence signal and the first timing relationship are input into a frequency determination model to obtain a second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the vital sign of the target object at the end moment of the second time window, thereby achieving the purpose of more accurately obtaining the real-time frequency of the user's vital signs, thereby achieving the technical effect of improving the accuracy when using an optical fiber sensor to monitor the frequency of the user's vital signs, and further solving the technical problem of difficulty in obtaining accurate monitoring results when using an optical fiber sensor to monitor the frequency of the user's vital signs. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0018] Figure 1 A hardware structure block diagram of a computer terminal for implementing a frequency determination method is shown;

[0019] Figure 2 is a flow chart of a frequency determination method according to an embodiment of the present invention;

[0020] Figure 3 is a schematic diagram of an application flow of a frequency determination model provided in an optional embodiment of the present invention;

[0021] Figure 4 is a schematic diagram of a training process of a frequency determination model provided according to an optional embodiment of the present invention;

[0022] Figure 5 is a structural block diagram of a frequency determination device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] According to an embodiment of the present invention, an embodiment of a frequency determination method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0026] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 FIG1 shows a hardware structure block diagram of a computer terminal for implementing a frequency determination method. Figure 1 As shown, the computer terminal 10 may include one or more processors (processors 102a, 102b, ..., 102n are used as examples in the figure) (the processors may include but are not limited to processing devices such as microprocessors MCU or programmable logic devices FPGA), and a memory 104 for storing data. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply, and / or a camera. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0027] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry." The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuitry may be a single, independent processing module, or may be incorporated in whole or in part into any of the other components of the computer terminal 10. As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0028] Memory 104 can be used to store software programs and modules for application software, such as the program instructions / data storage device corresponding to the frequency determination method in the embodiments of the present invention. The processor executes the software programs and modules stored in memory 104 to perform various functional applications and data processing, thereby implementing the frequency determination method for the application described above. Memory 104 can include high-speed random access memory (RAM) and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 104 can further include memory remotely located from the processor, and such remote memory can be connected to computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0029] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 .

[0030] In order to solve the problem of monitoring result accuracy when using optical fiber sensors to monitor the frequency of a user's vital signs, the present application provides the following embodiment, which uses a deep learning model to determine the frequency of the user's vital signs at the current moment.

[0031] Fiber optic sensors can be placed next to the user to monitor the user's body surface vibration signals. A fiber optic sensor is a sensor that uses optical fiber as a sensing element and uses the physical properties of light when propagating in the optical fiber to detect and measure changes in physical quantities. It can collect optical power sequence signals in real time. A fiber optic sensor is a sensor that converts the state of the object being measured into a measurable optical signal. It consists of an amplifier (control center) and an optical fiber. The fiber optic sensor has the characteristics of high sensitivity, resistance to electromagnetic interference, and high resolution. The optical power sequence signal is one of the signals collected by the fiber optic sensor, which can represent the power changes during the optical fiber transmission process. In the optical fiber communication system, the optical power sequence signal can reflect the quality and status of the optical fiber transmission, such as the optical fiber loss, pulse broadening and other parameters.

[0032] Figure 2FIG. 1 is a flow chart of a frequency determination method according to an embodiment of the present invention. Figure 2 As shown, the method includes the following steps:

[0033] Step S202 : Acquire a first frequency value of the target object, wherein the first frequency value represents the frequency of the vital sign of the target object at the end time of the first time window.

[0034] Wherein, the target object can be the user to be monitored, and the purpose of this embodiment is to monitor the real-time vital sign frequency of the target object. The first time window can be the time window corresponding to the previous moment, and the end time of the first time window is the previous moment, and the end time of the second time window is the current moment. The vital signs in this embodiment include vital signs that produce regular vibrations in the human body. Such vital signs can generate surface vibration signals that can be analyzed from the optical power sequence signal of the optical fiber sensor, so the optical fiber sensor can be used for monitoring. As an optional embodiment, the vital signs can be breathing or heartbeat, and the frequency of the vital signs can be the respiratory rate and heart rate of the target object. Breathing and heartbeat both produce regular vibrations on the human body surface, so the optical power sequence signal including the vibration component can be collected by the optical fiber sensor, and the respiratory rate or heart rate of the human body can be analyzed therefrom.

[0035] As an optional embodiment, the first frequency value of the target object can be obtained in the following manner: obtaining a prior optical power sequence signal, wherein the prior optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a first time window; performing a fast Fourier transform on the prior optical power sequence signal to obtain a prior frequency domain signal; and determining the frequency corresponding to the maximum amplitude in the prior frequency domain signal as the first frequency value.

[0036] The frequency determination model provided in this embodiment requires a pre-frequency value to represent the target subject's vital sign frequency at the previous moment before the predicted moment. For example, when the predicted moment is the current moment, the pre-frequency value represents the target subject's vital sign frequency at the previous moment. Based on this pre-frequency value, the frequency determination model predicts whether the target subject's vital sign frequency at the current moment has increased or decreased relative to the previous moment, thereby predicting the vital sign frequency at the current moment. This optional embodiment provides a method for obtaining a first frequency value, using a fast Fourier transform to estimate the first frequency value corresponding to a first time window. The fast Fourier transform (FFT) is an efficient algorithm for calculating the discrete Fourier transform (DFT) and its inverse transform. The basic idea of ​​the FFT is to decompose the original N-point sequence into a series of short sequences. By fully utilizing the symmetric and periodic properties of the exponential factors in the DFT calculation formula, the corresponding DFTs of these short sequences are calculated and appropriately combined to eliminate duplicate calculations, reduce multiplication operations, and simplify the structure. Through fast Fourier transform, the prior optical power sequence signal (time domain signal) can be converted into a frequency domain signal, and the frequency corresponding to the maximum amplitude in the frequency domain signal is considered to be the frequency of the vital sign, thereby achieving a brief estimation of the first frequency value. This method is simple and fast to calculate.

[0037] Step S204 , obtaining an optical power sequence signal, wherein the optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a second time window, and a first timing relationship between the first time window and the second time window is that the second time window is later than the first time window.

[0038] Optionally, the second time window and the first time window can have the same window duration. For example, if the time range covered by the first time window is recorded as the 1st second to the tth second, and if the second time window is 1 second later than the first time window, the time range covered by the second time window can be recorded as the 2nd second to the t+1th second. The purpose of this embodiment is to determine the target subject's vital sign frequency at the end time of the second time window, and the end time of the second time window can be the current time of real-time monitoring.

[0039] As an optional embodiment, before obtaining the optical power sequence signal, the following steps are also included to generate a second time window: obtaining the vital signs update cycle; according to the vital signs update cycle, shifting the first time window backward to obtain a second time window, wherein the time difference between the second time window and the first time window matches the vital signs update cycle.

[0040] Based on this optional embodiment, the update period of the vital signs frequency can be adjusted, and the accuracy of the vital signs frequency output by the frequency determination model can also be adjusted. Let the vital signs update period be T, and the time range covered by the first time window be recorded as the 1st second to the tth second, then the coverage range of the second time window can be determined as the 1st + Tth second to the tth + Tth second, that is, the first time window is shifted back by T seconds as a whole. Optionally, the window duration of the first time window is t seconds, then the T value range as the vital signs update period is (0, t), that is, greater than zero and less than t. The purpose of such a value is to allow a correlation between the optical power sequence signal in the second time window and the first frequency value determined according to the first time window. It is understandable that if there is no overlap between the first time window and the second time window, or even if the first time window and the second time window are far apart, then the frequency of the vital signs of the target object at the end of the first time window has almost no effect on the signal waveform of the optical power sequence signal in the second time window. The optical power sequence signal in the second time window cannot be used to predict the rising trend or falling trend of the frequency of the vital signs at the end of the second time window relative to the end of the first time window, resulting in the frequency determination model being unable to accurately predict the second frequency value.

[0041] Therefore, the greater the overlap between the first and second time windows, the closer the correlation between the optical power sequence signal corresponding to the second time window and the first frequency value, and the higher the accuracy of predicting the second frequency value using the optical power sequence signal corresponding to the second time window and the first frequency value. Optionally, an accuracy parameter can be defined, where the ratio α of the overlap between the first and second time windows and the window length is determined as the accuracy parameter, where α = T / t. The target user can freely adjust the accuracy parameter to achieve a balance between timeliness and accuracy in vital sign frequency updates.

[0042] Step S206: Input the first frequency value, the optical power sequence signal, and the first timing relationship into a frequency determination model to obtain a second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the vital sign of the target object at the end of the second time window.

[0043] The frequency determination model can be a deep learning model, such as a convolutional neural network model. Optionally, the frequency determination model can include convolutional layers, batch normalization layers, activation layers, pooling layers, and fully connected layers. The model's inputs are the vital sign frequency values ​​corresponding to the previous moment and the optical power sequence signal corresponding to the current moment. The convolutional layer is the core component of the convolutional neural network, responsible for extracting features from the input data. The convolutional layer captures local features by sliding a filter (or convolution kernel) over the input data and performing a convolution operation. The batch normalization layer is used to accelerate network training and improve model stability. By normalizing the mean and variance of each batch of data, the batch normalization layer stabilizes the network's input distribution, thereby improving training convergence. The activation layer introduces nonlinear factors, enabling the neural network to better learn and represent complex patterns. It serves to transfer nonlinear features between the convolutional and fully connected layers. The pooling layer, located after the convolutional layer, downsamples features, reducing the data dimension and improving the model's generalization ability. This effectively reduces the number of model parameters and helps prevent overfitting. The fully connected layer is responsible for combining learned features to predict the second frequency value. Each node in the fully connected layer is connected to all nodes in the previous layer. By combining all feature information, the fully connected layer achieves the final prediction.

[0044] As an optional embodiment, the frequency determination model is a model generated in advance through the following steps: obtaining training samples, wherein the training samples include: a first label representing the vital sign frequency of the sample object at a first moment, a second label representing the vital sign frequency of the sample object at a second moment, and an optical power sequence signal sample; the first moment is the end moment of the first sample time window, the second moment is the end moment of the second sample time window, and the optical power sequence signal sample is obtained by the optical fiber sensor monitoring the sample object within the second sample time window; the first label and the optical power sequence signal sample are input into the original model corresponding to the frequency determination model to obtain the vital sign frequency prediction value output by the original model; according to the difference between the vital sign frequency prediction value and the second label, the model parameters of the original model are updated to obtain the frequency determination model.

[0045] Optionally, the original model can be trained through multiple iterations, with the training samples including multiple groups of samples. The model is iteratively trained using these multiple groups of training samples until a predetermined number of iterations is reached, or until the model training results converge to a satisfactory accuracy. Model training is terminated to obtain a frequency determination model. The frequency determination model trained in this manner has higher accuracy and is better suited to the task of vital sign frequency prediction.

[0046] As an optional embodiment, the first frequency value, the optical power sequence signal and the first timing relationship are input into the frequency determination model to obtain the second frequency value output by the frequency determination model, including the following steps: inputting the optical power sequence signal into the feature extraction layer in the frequency determination model to obtain the frequency characteristics output by the feature extraction layer; inputting the first timing relationship and the frequency characteristics into the splicing layer in the frequency determination model to obtain the splicing vector output by the splicing layer, wherein the splicing vector is obtained by splicing the frequency characteristics and the first frequency value according to the first timing relationship; inputting the splicing vector into the prediction layer in the frequency determination model to obtain the second frequency value output by the prediction layer.

[0047] Optionally, the feature extraction layer, concatenation layer, and prediction layer can be descriptions of the various layers in the frequency determination model according to their respective functions. For example, the feature extraction layer can be implemented as a convolutional layer, and the prediction layer can be implemented as a fully connected layer. Based on this optional embodiment, the logical process of the frequency determination model predicting the second frequency value is described, which provides greater explainability for the prediction of the second frequency value and facilitates adjustment of model parameters and improvement of model training.

[0048] As an optional embodiment, after obtaining the second frequency value output by the frequency determination model, the method also includes: obtaining a subsequent optical power sequence signal, wherein the subsequent optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a third time window, and the second timing relationship between the second time window and the third time window is that the third time window is later than the second time window; inputting the second frequency value, the subsequent optical power sequence signal and the second timing relationship into the frequency determination model to obtain a third frequency value output by the frequency determination model, wherein the third frequency value represents the frequency of the vital signs of the target object at the end moment of the third time window.

[0049] Based on this optional embodiment, the frequency determination model can be used to predict the vital sign frequency at the next moment based on the current moment's vital sign frequency prediction results, achieving continuous prediction of vital sign frequencies. The end time of the second time window is the current moment, and the end time of the third time window can be considered the next moment. Subsequently, the method provided in this optional embodiment can be used to continuously collect the target object's optical power sequence signal, providing the target object with continuous vital sign frequency prediction results.

[0050] In the above steps, a first frequency value of the target object is obtained, wherein the first frequency value represents the frequency of the vital signs of the target object at the end of the first time window; an optical power sequence signal is obtained, wherein the optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within the second time window, and the first timing relationship between the first time window and the second time window is that the second time window is later than the first time window; the first frequency value, the optical power sequence signal and the first timing relationship are input into the frequency determination model to obtain the second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the vital signs of the target object at the end of the second time window, thereby achieving the purpose of more accurately obtaining the real-time frequency of the user's vital signs, thereby achieving the technical effect of improving the accuracy when using optical fiber sensors to monitor the frequency of user vital signs, and further solving the technical problem of difficulty in obtaining accurate monitoring results when using optical fiber sensors to monitor the frequency of user vital signs.

[0051] Figure 3 FIG. 1 is a schematic diagram of an application flow of a frequency determination model provided in an optional embodiment of the present invention. Figure 3 As shown, the frequency determination model in this optional embodiment can predict the user's current heart rate or breathing rate through the following steps:

[0052] Use an optical fiber sensor to collect the user's optical power sequence signal in real time. After collecting the signal for t seconds, use the FFT method to calculate the FFT heart rate label / respiration label corresponding to the tth second. When collecting the optical power sequence signal to t+1 second, calculate the heart rate / respiration time domain signal from the 2nd to t+1 seconds, and use the FFT method to obtain the FFT heart rate / respiration label at the current moment. At the same time, the time domain signal from the 2nd to t+1 seconds and the FFT heart rate label / respiration label at the previous moment are fed into the frequency determination model to obtain the predicted heart rate / respiration rate corresponding to the t+1 second. Similarly, the predicted heart rate / respiration rate for the t+2th, t+3th, and subsequent moments can be obtained in the same way, that is, after t seconds, the heart rate / respiration rate per second can be obtained.

[0053] Figure 4 FIG. 1 is a schematic diagram of a training process of a frequency determination model according to an optional embodiment of the present invention. Figure 4 As shown, the frequency determination model can be trained through the following steps:

[0054] Step 1: Obtain the FFT heart rate label / respiration label of the previous moment and the heart rate time domain signal / respiration time domain signal of the current moment. The heart rate time domain signal / respiration time domain signal of the current moment can be obtained through the following steps.

[0055] Based on the optical power sequence signal corresponding to the time window at the current moment collected by the fiber optic sensor, the acquisition frequency is recorded as sample_fre. The fixed duration of the time window is t seconds. There are a total of sample_fre*t points in this time window. In this case, the heart rate / respiratory rate collected by the ECG monitor can be used as the label. This optical power sequence signal is input into a filter with a frequency bandwidth corresponding to the heart rate for filtering. The heart rate filtered value, i.e., the heart rate time domain signal, is then converted into a frequency domain signal through a fast Fourier transform (FFT). The FFT heart rate label is obtained based on the frequency corresponding to the maximum amplitude of the frequency domain signal.

[0056] Similarly, the optical power sequence signal can be input into a filter with a frequency bandwidth range corresponding to breathing for filtering processing to obtain a respiratory filter value, i.e., a respiratory time domain signal, which is then converted into a frequency domain signal by fast Fourier transform (FFT), and the FFT respiratory label is obtained according to the frequency corresponding to the maximum amplitude of the frequency domain signal. When the window shift is 1 second, the FFT heart rate / respiration label of the previous moment and the heart rate / respiration time domain signal of the current moment can be obtained. For example, the optical power sequence from the 1st second to the tth second passes through a filter with a required frequency bandwidth range to obtain a heart rate / respiration time domain signal, and then obtains an FFT heart rate / respiration label through FFT, which is recorded as the FFT heart rate / respiration label of the previous moment; then the heart rate / respiration time domain signal obtained by the optical power sequence from the 2nd second to the t+1 second through a filter with a required frequency bandwidth range is the heart rate / respiration time domain signal of the current moment.

[0057] Step 2: The model extracts features and concatenates them with the FFT heart rate / respiration at the previous moment.

[0058] Taking the training process of the frequency prediction model corresponding to heart rate as an example, set the training cycle to T and record the original model f θ =f n f n-1 f n-2 ...f 1 (·) This model is a convolutional neural network consisting of convolutional layers, batch normalization (BN) layers, activation layers, pooling layers, and fully connected layers. The current heart rate time domain signal is input into the original model. The first m layers extract the heart rate features and concatenate them with the FFT heart rate label from the previous moment. This concatenated vector is then fed into the model. The model then learns the upward or downward trend of the current heart rate relative to the previous heart rate in the last nm layers, outputting the predicted heart rate. Similarly, the respiration model outputs the predicted respiration.

[0059] Step 3: Optimize training and update the model.

[0060] The error between the heart rate / respiration prediction value and the heart rate / respiration label is calculated using the loss function, and the heart rate / respiration model parameters are updated through backpropagation.

[0061] Step 4: Completing steps 2 to 3 means completing one cycle of training. Repeat steps 2 to 3 until the training is completed.

[0062] It should be noted that for the aforementioned method embodiments, for simplicity of description, they are all expressed as a series of action combinations. However, those skilled in the art should be aware that the present invention is not limited by the order of the actions described, because according to the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0063] Through the description of the above embodiments, those skilled in the art can clearly understand that the frequency determination method according to the above embodiment can be implemented by software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a more preferred embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0064] According to an embodiment of the present invention, a frequency determination device for implementing the above-mentioned frequency determination method is also provided. Figure 5 is a structural block diagram of a frequency determination device provided according to an embodiment of the present invention. Figure 5 As shown, the frequency determination device includes: a first acquisition module 52, a second acquisition module 54 and a determination module 56. The frequency determination device is described below.

[0065] A first acquisition module 52 is configured to acquire a first frequency value of the target object, wherein the first frequency value represents the frequency of the target object's vital sign at the end of the first time window;

[0066] A second acquisition module 54 is connected to the first acquisition module 52 and is used to acquire an optical power sequence signal, wherein the optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a second time window, and a first timing relationship between the first time window and the second time window is that the second time window is later than the first time window;

[0067] The determination module 56 is connected to the above-mentioned second acquisition module 54, and is used to input the first frequency value, the optical power sequence signal and the first timing relationship into the frequency determination model to obtain a second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the vital signs of the target object at the end time of the second time window.

[0068] It should be noted that the first acquisition module 52, the second acquisition module 54, and the determination module 56 correspond to steps S202 to S206 in the embodiment. The examples and application scenarios implemented by the three modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above embodiment. It should be noted that the above modules, as part of the device, can be run in the computer terminal 10 provided in the embodiment.

[0069] An embodiment of the present invention may provide a computer device. Optionally, in this embodiment, the computer device may be located in at least one of a plurality of network devices in a computer network. The computer device includes a memory and a processor.

[0070] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the frequency determination method and device in the embodiments of the present invention. The processor executes the software programs and modules stored in the memory to perform various functional applications and data processing, thereby implementing the above-mentioned frequency determination method. The memory may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include memory remotely located relative to the processor, and these remote memories may be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0071] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: obtaining a first frequency value of the target object, wherein the first frequency value represents the frequency of the vital signs of the target object at the end of the first time window; obtaining an optical power sequence signal, wherein the optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within the second time window, and the first timing relationship between the first time window and the second time window is that the second time window is later than the first time window; inputting the first frequency value, the optical power sequence signal and the first timing relationship into the frequency determination model to obtain a second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the vital signs of the target object at the end of the second time window.

[0072] Optionally, the processor may also execute the following program code: inputting the first frequency value, the optical power sequence signal and the first timing relationship into the frequency determination model to obtain the second frequency value output by the frequency determination model, including: inputting the optical power sequence signal into the feature extraction layer in the frequency determination model to obtain the frequency characteristics output by the feature extraction layer; inputting the first timing relationship and the frequency characteristics into the splicing layer in the frequency determination model to obtain the splicing vector output by the splicing layer, wherein the splicing vector is obtained by splicing the frequency characteristics and the first frequency value according to the first timing relationship; inputting the splicing vector into the prediction layer in the frequency determination model to obtain the second frequency value output by the prediction layer.

[0073] Optionally, the processor may also execute the program code of the following steps: the frequency determination model is a model generated in advance through the following steps: obtaining a training sample, wherein the training sample includes: a first label representing the vital sign frequency of the sample object at a first moment, a second label representing the vital sign frequency of the sample object at a second moment, and an optical power sequence signal sample; the first moment is the end moment of the first sample time window, the second moment is the end moment of the second sample time window, and the optical power sequence signal sample is obtained by the optical fiber sensor monitoring the sample object within the second sample time window; the first label and the optical power sequence signal sample are input into the original model corresponding to the frequency determination model to obtain the vital sign frequency prediction value output by the original model; according to the difference between the vital sign frequency prediction value and the second label, the model parameters of the original model are updated to obtain the frequency determination model.

[0074] Optionally, the above-mentioned processor can also execute the program code of the following steps: before obtaining the optical power sequence signal, it also includes: obtaining the vital signs update cycle; according to the vital signs update cycle, shifting the first time window backward to obtain a second time window, wherein the time difference between the second time window and the first time window matches the vital signs update cycle.

[0075] Optionally, the processor may also execute program code for the following steps: obtaining a first frequency value of the target object, including: obtaining a prior optical power sequence signal, wherein the prior optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a first time window; performing a fast Fourier transform on the prior optical power sequence signal to obtain a prior frequency domain signal; and determining the frequency corresponding to the maximum amplitude in the prior frequency domain signal as the first frequency value.

[0076] Optionally, the processor may also execute the program code of the following steps: after obtaining the second frequency value output by the frequency determination model, the method further includes: obtaining a subsequent optical power sequence signal, wherein the subsequent optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a third time window, and the second timing relationship between the second time window and the third time window is that the third time window is later than the second time window; inputting the second frequency value, the subsequent optical power sequence signal and the second timing relationship into the frequency determination model to obtain a third frequency value output by the frequency determination model, wherein the third frequency value represents the frequency of the target object's vital signs at the end moment of the third time window.

[0077] Optionally, the processor may further execute program code of the following steps: the vital sign is breathing or heartbeat.

[0078] A person skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a non-volatile storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0079] The embodiment of the present invention further provides a non-volatile storage medium. Optionally, in this embodiment, the non-volatile storage medium can be used to store the program code executed by the frequency determination method provided in the above embodiment.

[0080] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0081] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: obtaining a first frequency value of the target object, wherein the first frequency value represents the frequency of the vital signs of the target object at the end of the first time window; obtaining an optical power sequence signal, wherein the optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within the second time window, and the first timing relationship between the first time window and the second time window is that the second time window is later than the first time window; inputting the first frequency value, the optical power sequence signal and the first timing relationship into a frequency determination model to obtain a second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the vital signs of the target object at the end of the second time window.

[0082] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: inputting the first frequency value, the optical power sequence signal and the first timing relationship into the frequency determination model to obtain the second frequency value output by the frequency determination model, including: inputting the optical power sequence signal into the feature extraction layer in the frequency determination model to obtain the frequency characteristics output by the feature extraction layer; inputting the first timing relationship and the frequency characteristics into the splicing layer in the frequency determination model to obtain the splicing vector output by the splicing layer, wherein the splicing vector is obtained by splicing the frequency characteristics and the first frequency value according to the first timing relationship; inputting the splicing vector into the prediction layer in the frequency determination model to obtain the second frequency value output by the prediction layer.

[0083] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: the frequency determination model is a model generated in advance by training through the following steps: obtaining a training sample, wherein the training sample includes: a first label representing the vital sign frequency of the sample object at a first moment, a second label representing the vital sign frequency of the sample object at a second moment, and an optical power sequence signal sample; the first moment is the end moment of the first sample time window, the second moment is the end moment of the second sample time window, and the optical power sequence signal sample is obtained by the optical fiber sensor monitoring the sample object within the second sample time window; the first label and the optical power sequence signal sample are input into the original model corresponding to the frequency determination model to obtain the vital sign frequency prediction value output by the original model; according to the difference between the vital sign frequency prediction value and the second label, the model parameters of the original model are updated to obtain the frequency determination model.

[0084] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: before obtaining the optical power sequence signal, it also includes: obtaining the vital signs update cycle; according to the vital signs update cycle, shifting the first time window backward to obtain a second time window, wherein the time difference between the second time window and the first time window matches the vital signs update cycle.

[0085] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: obtaining a first frequency value of the target object, including: obtaining a prior optical power sequence signal, wherein the prior optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a first time window; performing a fast Fourier transform on the prior optical power sequence signal to obtain a prior frequency domain signal; and determining the frequency corresponding to the maximum amplitude in the prior frequency domain signal as the first frequency value.

[0086] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: after obtaining the second frequency value output by the frequency determination model, the method further includes: obtaining a subsequent optical power sequence signal, wherein the subsequent optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a third time window, and the second timing relationship between the second time window and the third time window is that the third time window is later than the second time window; inputting the second frequency value, the subsequent optical power sequence signal and the second timing relationship into the frequency determination model to obtain a third frequency value output by the frequency determination model, wherein the third frequency value represents the frequency of the vital signs of the target object at the end moment of the third time window.

[0087] Optionally, in this embodiment, the non-volatile storage medium is configured to store program codes for executing the following steps: the vital sign is breathing or heartbeat.

[0088] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0089] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0090] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0091] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected to achieve the purpose of the present embodiment according to actual needs.

[0092] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0093] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, and other media that can store program codes.

[0094] The above are only preferred embodiments of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A frequency determination method, characterized in that: include: Acquire a first frequency value of the target object, wherein the first frequency value represents the frequency of the vital sign of the target object at the end time of the first time window; Acquire an optical power sequence signal, wherein the optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a second time window, and a first timing relationship between the first time window and the second time window is that the second time window is later than the first time window; The first frequency value, the optical power sequence signal, and the first timing relationship are input into a frequency determination model to obtain a second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the vital sign of the target object at the end moment of the second time window.

2. The method according to claim 1, characterized in that Inputting the first frequency value, the optical power sequence signal, and the first timing relationship into a frequency determination model to obtain a second frequency value output by the frequency determination model includes: Inputting the optical power sequence signal into the feature extraction layer in the frequency determination model to obtain the frequency characteristics output by the feature extraction layer; Inputting the first time series relationship and the frequency feature into a splicing layer in the frequency determination model to obtain a splicing vector output by the splicing layer, wherein the splicing vector is obtained by splicing the frequency feature and the first frequency value according to the first time series relationship; The splicing vector is input into a prediction layer in the frequency determination model to obtain the second frequency value output by the prediction layer.

3. The method according to claim 1, characterized in that The frequency determination model is a model generated by pre-training through the following steps: Acquire a training sample, wherein the training sample includes: a first label representing the frequency of a vital sign of a sample subject at a first moment, a second label representing the frequency of a vital sign of the sample subject at a second moment, and an optical power sequence signal sample; the first moment is the end moment of a first sample time window, the second moment is the end moment of a second sample time window, and the optical power sequence signal sample is obtained by the optical fiber sensor monitoring the sample subject within the second sample time window; Inputting the first label and the optical power sequence signal sample into an original model corresponding to the frequency determination model to obtain a vital sign frequency prediction value output by the original model; According to the difference between the vital sign frequency prediction value and the second label, the model parameters of the original model are updated to obtain the frequency determination model.

4. The method according to claim 1, wherein Before acquiring the optical power sequence signal, the method further includes: Get vital signs update cycle; According to the vital sign update cycle, the first time window is shifted back to obtain the second time window, wherein the time difference between the second time window and the first time window matches the vital sign update cycle.

5. The method according to claim 1, wherein The obtaining of the first frequency value of the target object includes: Acquiring a prior optical power sequence signal, wherein the prior optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within the first time window; Performing a fast Fourier transform on the prior optical power sequence signal to obtain a prior frequency domain signal; The frequency corresponding to the maximum amplitude value in the previous frequency domain signal is determined as the first frequency value.

6. The method according to claim 1, characterized in that After obtaining the second frequency value output by the frequency determination model, the method further includes: Acquire a subsequent optical work sequence signal, wherein the subsequent optical work sequence signal is obtained by the optical fiber sensor monitoring the target object within a third time window, and a second timing relationship between the second time window and the third time window is that the third time window is later than the second time window; The second frequency value, the subsequent optical power sequence signal and the second timing relationship are input into the frequency determination model to obtain a third frequency value output by the frequency determination model, wherein the third frequency value represents the frequency of the vital sign of the target object at the end time of the third time window.

7. The method according to any one of claims 1 to 6, characterized in that The vital sign is breathing or heartbeat.

8. A frequency determination device, characterized in that: include: A first acquisition module is configured to acquire a first frequency value of a target object, wherein the first frequency value represents a frequency of a vital sign of the target object at an end time of a first time window; a second acquisition module, configured to acquire an optical power sequence signal, wherein the optical power sequence signal is obtained by the optical fiber sensor monitoring the target object within a second time window, and a first timing relationship between the first time window and the second time window is that the second time window is later than the first time window; A determination module is used to input the first frequency value, the optical power sequence signal and the first timing relationship into a frequency determination model to obtain a second frequency value output by the frequency determination model, wherein the second frequency value represents the frequency of the vital sign of the target object at the end time of the second time window.

9. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute the frequency determination method according to any one of claims 1 to 7.

10. A computer device, characterized in that: The computer device includes a memory and a processor, the memory is used to store a program, and the processor is used to run the program stored in the memory, wherein the frequency determination method according to any one of claims 1 to 7 is executed when the program is run.

Images