Identity recognition method and system based on pressure sensor, medium and equipment

Abstract

The invention discloses an identity recognition method and system based on a pressure sensor, a medium and equipment, and relates to the technical field of identity recognition. The method comprises the steps that a pressure signal, collected by a pressure sensor, of a target person is obtained, and the pressure signal is converted into heart rate data; performing data preprocessing on the heart rate data to obtain an effective signal segment; extracting waveform feature parameters in the effective signal segments, and performing feature vector conversion on the waveform feature parameters according to a preset feature model to obtain feature vectors; and matching the feature vector with a preset authorized user template to obtain an identity verification result of the target person. By implementing the technical scheme provided by the invention, the accuracy of identity recognition can be improved.

Description

Technical Field

[0001] The present application relates to the field of identity recognition technology, and in particular to an identity recognition method, system, medium and device based on a pressure sensor. Background Art

[0002] With the advancement of science and technology and the development of society, the demand for personal privacy protection is increasing. As an efficient and secure technology, biometric recognition has been widely used in access control systems, smart phones, personal computers, etc. Traditional biometric recognition mainly includes fingerprint recognition, iris scanning, facial recognition, etc. Although these technologies have certain security and convenience, they still have limitations in specific application scenarios. For example, the recognition effect will be significantly affected when wearing gloves or in poor ambient light.

[0003] At present, in order to overcome the limitations of traditional methods, researchers have proposed identity recognition methods based on physiological signals. Among them, heart rate signals, as one of the important physiological characteristics of the human body, have gradually become an important research direction for identity recognition due to their uniqueness and stability. Among them, the method of using the human body as a conductor for electrode detection has gradually gained attention, such as measuring changes in human skin resistance or electrocardiogram (ECG) signals for individual recognition. However, such methods usually require direct contact with human skin and are easily affected by fluctuations in human body state, resulting in low accuracy of identity recognition. Summary of the invention

[0004] The present application provides an identity recognition method based on pressure sensing, which can improve the accuracy of identity recognition.

[0005] In a first aspect, the present application provides an identity recognition method based on pressure sensing, the method comprising: Acquire a pressure signal of a target person collected by a pressure sensor, and convert the pressure signal into heart rate data; Performing data preprocessing on the heart rate data to obtain a valid signal segment; Extracting waveform characteristic parameters from the effective signal segment, and performing feature vector conversion on the waveform characteristic parameters according to a preset feature model to obtain a feature vector; The feature vector is matched with a preset authorized user template to obtain the identity authentication result of the target person.

[0006] By adopting the above technical solution, the pressure signal of the target person is collected by a pressure sensor and converted into heart rate data, which avoids the problem of direct contact with human skin in traditional physiological signal recognition, improves the convenience of the recognition process and user experience, and does not need to consider the influence of external factors such as ambient light. It overcomes the application limitations of traditional fingerprint recognition and facial recognition in specific scenarios. By preprocessing the heart rate data to obtain an effective signal segment, it can effectively filter out noise and interference in the signal acquisition process and improve the accuracy of subsequent feature extraction. In the feature extraction link, this solution extracts the waveform feature parameters in the effective signal segment and converts the feature vector in combination with the preset feature model, thereby realizing efficient mapping from the original heart rate data to the feature vector. Finally, the identity authentication result is determined by matching the feature vector with the preset authorized user template, thereby realizing effective extraction and reliable identification of individual features in the heart rate signal, thereby improving the accuracy of identity recognition.

[0007] Optionally, converting the pressure signal into heart rate data includes: Performing charge amplification on the pressure signal to obtain an amplified voltage signal; Performing analog-to-digital conversion on the amplified voltage signal to obtain a digital quantity signal; Calculating peak values ​​and trough values ​​according to the digital signal, and determining a heart rate cycle based on the peak values ​​and trough values; Heart rate data is generated according to the heart rate cycle.

[0008] Optionally, the performing data preprocessing on the heart rate data to obtain a valid signal segment includes: Performing bandpass filtering on the heart rate data to obtain an initial filtered signal; Calculating a signal-to-noise ratio of the initial filtered signal, and setting a signal threshold according to the signal-to-noise ratio; The signal that is greater than the signal threshold is extracted as a valid signal segment.

[0009] Optionally, extracting waveform characteristic parameters in the valid signal segment includes: Calculating the waveform change rate in the effective signal segment, and identifying the starting point and the ending point of the dicrotic wave according to the waveform change rate; Calculate the area, peak value and duration of the dicrotic wave; Calculating a dicrotic wave index based on the area, peak value and duration; Waveform characteristic parameters are extracted based on the dicrotic wave index.

[0010] Optionally, performing feature vector conversion on the waveform feature parameter according to a preset feature model to obtain a feature vector includes: Inputting the waveform characteristic parameters into a preset characteristic model to obtain a characteristic correlation matrix; Performing feature clustering analysis based on the feature correlation matrix, and calculating weight coefficients of various features according to the clustering analysis results; Performing a weighted combination of the features based on the weight coefficients; The weighted combination result is subjected to dimensionality reduction processing to obtain a feature vector.

[0011] Optionally, matching the feature vector with a preset authorized user template to obtain the identity authentication result of the target person includes: Calculate a feature difference value between the feature vector and each template feature vector in a preset authorized user template; constructing a similarity score based on the feature difference values; Set dynamic matching thresholds based on historical matching data; When the similarity score is greater than the dynamic matching threshold, confirming that the identity authentication is passed; When the similarity score is not greater than the dynamic matching threshold, it is determined that the identity authentication has failed.

[0012] Optionally, after obtaining the pressure signal of the target person collected by the pressure sensor, the method further includes: monitoring the amplitude change rate of the pressure signal in real time, and determining it as an abnormal fluctuation when the amplitude change rate exceeds a preset range; Calculating the duration and amplitude of the abnormal fluctuation; When the duration is less than the time threshold and the fluctuation amplitude is less than the amplitude threshold, smoothing the abnormal fluctuation; When the duration is not less than the time threshold or the fluctuation amplitude is not less than the amplitude threshold, the currently collected pressure signal is marked as an invalid signal, and the pressure signal of the target person collected by the pressure sensor is reacquired until a valid pressure signal is collected.

[0013] In a second aspect of the present application, a pressure sensor-based identity recognition system is provided, the system comprising: A data acquisition module, used to acquire a pressure signal of a target person collected by a pressure sensor and convert the pressure signal into heart rate data; A data processing module, used for performing data preprocessing on the heart rate data to obtain a valid signal segment; A data conversion module, used for extracting waveform characteristic parameters in the effective signal segment, and performing characteristic vector conversion on the waveform characteristic parameters according to a preset characteristic model to obtain a characteristic vector; The data verification module is used to match the feature vector with a preset authorized user template to obtain the identity verification result of the target person.

[0014] In a third aspect of the present application, a computer storage medium is provided, wherein the computer storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the above method steps.

[0015] In a fourth aspect of the present application, an electronic device is provided, comprising: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above-mentioned method steps.

[0016] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: This application collects the pressure signal of the target person through a pressure sensor and converts it into heart rate data, avoiding the problem of direct contact with human skin in traditional physiological signal recognition, improving the convenience and user experience of the recognition process, and eliminating the need to consider the influence of external factors such as ambient light. It overcomes the application limitations of traditional fingerprint recognition and facial recognition in specific scenarios. By preprocessing the heart rate data to obtain an effective signal segment, it can effectively filter out noise and interference in the signal acquisition process and improve the accuracy of subsequent feature extraction. In the feature extraction stage, this solution extracts waveform feature parameters from the effective signal segment and converts the feature vector into a preset feature model, thereby achieving efficient mapping from raw heart rate data to feature vectors. Finally, the identity authentication result is determined by matching the feature vector with the preset authorized user template, thereby achieving effective extraction and reliable identification of individual features in the heart rate signal, thereby improving the accuracy of identity recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flow chart of an identity recognition method based on a pressure sensor provided in an embodiment of the present application; Figure 2 is a module schematic diagram of an identity recognition system based on a pressure sensor provided in an embodiment of the present application; Figure 3 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application.

[0018] Description of reference numerals: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0019] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0020] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0021] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0022] The following will provide a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0023] Please refer to Figure 1 , a flow chart of an identity recognition method based on a pressure sensor is proposed. The method can be implemented by a computer program, can be implemented by a single-chip microcomputer, or can be run on an identity recognition system based on a pressure sensor. The computer program can be integrated into a computer device or can be run as an independent tool application. Specifically, the method includes steps 10 to 40, and the steps are as follows: Step 10: Obtain the pressure signal of the target person collected by the pressure sensor, and convert the pressure signal into heart rate data.

[0024] The embodiments of the present application can be applied to various identity recognition scenarios, including but not limited to driver identity recognition of smart vehicles, identity recognition of smart homes, identity recognition of office equipment, and unlocking recognition of smart phones.

[0025] Specifically, the pressure signal of the target person is collected by a pressure sensor. The pressure sensor can be set on the contact surface of the device (such as the surface of the steering wheel, the surface of the seat, etc.). When the target person is in pressure contact with the contact surface, the pressure sensor can sense and record the pressure change. Since the heartbeat information of the human body is superimposed on the pressure signal when the human body is in pressure contact, the physiological characteristics related to the heart rate can be extracted from the pressure signal. Compared with traditional acquisition methods such as electrocardiogram (ECG) or photoelectric volume pulse wave (PPG), the acquisition of pressure signals is more natural and does not require the user to perform specific coordinated actions. Further, the pressure signal collected by the pressure sensor is amplified by a charge amplifier to obtain an amplified voltage signal. Since the original pressure signal is relatively weak, the signal strength can be improved by charge amplification, which is convenient for subsequent processing. Then, the amplified voltage signal is sampled and quantized by an analog-to-digital converter to convert the analog signal into a digital signal. The digital signal is convenient for computer processing and storage. Then, the digital signal is analyzed, the peak value and trough value of the signal are calculated, and the heart rate cycle is determined by the time interval between adjacent peaks or troughs. Finally, the heart rate data of the target person is calculated based on the heart rate cycle.

[0026] For example, in this embodiment, a pressure sensor can be integrated into the power button or screen frame of the smartphone. When the user presses the power button or holds the phone, the pressure sensor can collect the user's pressure signal. Because when the human finger or palm is in contact with the smartphone, the human heartbeat information is superimposed on the pressure signal. This pulsation information produces tiny pressure fluctuations as the rhythmic changes of blood circulation occur. Therefore, physiological characteristics related to heart rate can be extracted from the pressure signal. Compared with traditional fingerprint recognition or face recognition, this recognition method based on pressure signals does not require special alignment of the sensor, and can achieve reliable recognition even in dark environments or when the fingers are wet.

[0027] Based on the above embodiment, as an optional embodiment, the step of converting the pressure signal into heart rate data may further include the following steps: Step 101: Charge amplify the pressure signal to obtain an amplified voltage signal.

[0028] Specifically, the amplitude of the original pressure signal collected by the pressure sensor is small, usually in the microvolt (μV) level, which is easily interfered by environmental noise and is not conducive to subsequent signal processing. In order to improve the signal-to-noise ratio of the signal, the present embodiment uses a charge amplifier to amplify the pressure signal. Specifically, the charge amplifier includes an operational amplifier and a feedback capacitor, wherein the in-phase input terminal of the operational amplifier is grounded, the inverting input terminal is connected to the output terminal of the pressure sensor, and the feedback capacitor is connected in parallel between the input terminal and the output terminal of the operational amplifier. When the pressure sensor outputs a charge signal, the charge amplifier converts the input charge into a voltage signal proportional to it, and the amplification factor is determined by the capacitance of the feedback capacitor. By adjusting the capacitance of the feedback capacitor, the weak pressure signal can be amplified to the millivolt (mV) level to obtain an amplified voltage signal.

[0029] Step 102: Perform analog-to-digital conversion on the amplified voltage signal to obtain a digital signal.

[0030] Specifically, in order to facilitate the digital processing system to analyze the signal, it is necessary to convert the analog amplified voltage signal into a digital signal. This embodiment uses a built-in analog-to-digital converter to sample and quantize the amplified voltage signal. The reference voltage of the analog-to-digital converter is 3.3V, and the sampling accuracy is 12 bits, that is, the quantization level is 4096 levels, which can more accurately reflect the changes in the amplified voltage signal. At the same time, considering the frequency characteristics of the human heartbeat signal, the sampling frequency of the analog-to-digital converter is set to 1000Hz, which can not only ensure the integrity of the signal waveform, but also avoid excessive data redundancy. After charge amplification and analog-to-digital conversion, the original weak pressure signal is converted into a digital signal with appropriate amplitude and high quantization accuracy. The digital signal not only has a high signal-to-noise ratio, which is convenient for subsequent processing such as digital filtering, but also easy to store and transmit.

[0031] Step 103: Calculate the peak value and the trough value according to the digital signal, and determine the heart rate cycle based on the peak value and the trough value.

[0032] Specifically, since the digital signal may contain interference components such as high-frequency noise and baseline drift, directly extracting the heart rate feature may lead to misjudgment. In order to accurately identify the heartbeat cycle in the digital signal, the present embodiment first uses a bandpass filter to pre-process the digital signal. The passband range of the bandpass filter is set to 0.5Hz-4Hz, which covers the frequency range of normal human heartbeat (usually 0.67Hz-3Hz, corresponding to a heart rate of 40-180 times / minute), and can effectively filter out interference signals. The digital signal after filtering presents obvious periodic fluctuation characteristics. In order to accurately detect the peak value and trough value, the present embodiment uses a sliding window method for extreme value detection, and sets the window length to 1 / 10 of the number of sampling points. When the signal value of a sampling point in the window is greater than or less than the signal value of all other sampling points in the window, the sampling point is marked as a peak value point or a trough value point. At the same time, in order to avoid the interference of false extreme points, it is also necessary to judge the time interval between adjacent extreme points. If the interval is less than a preset threshold (for example, 0.25 seconds), it is considered that the detected point is a false extreme point caused by noise and is removed. After obtaining the peak value points and the trough value points, this embodiment determines the heart rate cycle by calculating the time interval between adjacent peak value points or between adjacent trough value points. Since there may be waveform distortion in the actual signal, this embodiment calculates the peak interval and the trough interval at the same time, and takes their average value as the heart rate cycle, which can improve the accuracy of the cycle measurement.

[0033] Step 104: Generate heart rate data according to the heart rate cycle.

[0034] Specifically, the heart rate data is calculated according to the heart rate cycle T, and the calculation formula of the heart rate value HR is: HR = 60 / T, in units of times / minute. In order to improve the reliability of the heart rate data, the present embodiment averages the heart rate values ​​calculated from a plurality of consecutive heart rate cycles to obtain the final heart rate data. Ten consecutive valid heart rate cycles can be selected for calculation, and the maximum and minimum values ​​can be eliminated and then averaged, so as to reduce the influence of accidental factors.

[0035] Based on the above embodiment, as an optional embodiment, after obtaining the pressure signal of the target person collected by the pressure sensor, the following process is also included: Specifically, since the user may experience hand tremors, sudden changes in pressure intensity, etc., the pressure signal collected by the pressure sensor may fluctuate abnormally, affecting the subsequent heart rate extraction and identity recognition. Therefore, after obtaining the pressure signal, it is first necessary to monitor the amplitude change rate of the pressure signal in real time. In this embodiment, the ratio of the pressure value change between adjacent sampling points to the sampling time interval is used to represent the amplitude change rate. Assuming that the pressure value of the current sampling point is P1, the pressure value of the previous sampling point is P2, and the sampling time interval is Δt, the amplitude change rate R = |P1-P2| / Δt. When the amplitude change rate R exceeds the preset range, that is, R>Rmax or R<Rmin, the system determines that abnormal fluctuations occur. Among them, the upper and lower limits Rmax and Rmin of the preset range can be obtained through experimental statistics. In this embodiment, Rmax is set to 200% / second of the normal pressure value, and Rmin is set to 10% / second of the normal pressure value.

[0036] When an abnormal fluctuation is detected, the duration T and fluctuation amplitude A of the abnormal fluctuation will be calculated. The duration T is obtained by accumulating the number of continuous sampling points whose amplitude change rate exceeds the preset range and multiplying it by the sampling time interval. The fluctuation amplitude A is calculated by the difference between the maximum pressure value and the minimum pressure value during the abnormal fluctuation. According to the analysis of experimental data, the time threshold Tth is set to 0.5 seconds, and the amplitude threshold Ath is set to 50% of the normal pressure value. For abnormal fluctuations with a duration T less than the time threshold Tth and a fluctuation amplitude A less than the amplitude threshold Ath, considering that such short-term and small-amplitude fluctuations may be caused by slight hand shaking and will not have a significant impact on the heart rate characteristics, a smoothing method is used to correct the abnormal fluctuations. Specifically, a moving average filter is used to smooth the pressure value in the abnormal fluctuation interval, and the filter window length is set to the number of sampling points within the abnormal duration, so that the abnormal fluctuations can be eliminated while retaining the main characteristics of the signal. For abnormal fluctuations with duration T not less than the time threshold Tth or fluctuation amplitude A not less than the amplitude threshold Ath, since such violent or continuous fluctuations may be caused by changes in the user's holding posture, shaking of the phone, etc., which will seriously affect the extraction of heart rate features, the currently collected pressure signal is marked as an invalid signal. The system will prompt the user to keep the phone stable and restart the pressure signal collection process until a valid pressure signal that meets the requirements is obtained.

[0037] Step 20: Preprocess the heart rate data to obtain valid signal segments.

[0038] Specifically, in this embodiment, since the heart rate data may be affected by factors such as the user's activity status and environmental interference, and abnormal values ​​or discontinuities may occur, in order to improve the accuracy of subsequent identity recognition, the heart rate data needs to be preprocessed to filter out valid signal segments. The purpose of preprocessing is to remove abnormal data and extract data segments that can truly reflect the user's heart rate characteristics. Heart rate data preprocessing refers to the process of noise reduction and signal quality evaluation of the collected raw heart rate data, and finally obtains valid signal segments.

[0039] Based on the above embodiment, as an optional embodiment, the step of performing data preprocessing on the heart rate data to obtain a valid signal segment may further include the following steps: Step 201: band-pass filter the heart rate data to obtain an initial filtered signal.

[0040] Specifically, since the heart rate data may contain non-heart rate components such as environmental noise, baseline drift and high-frequency interference, it is necessary to purify the signal through bandpass filtering. This embodiment uses a Butterworth bandpass filter to process the heart rate data. The filter has good frequency selectivity and phase characteristics. Considering that the frequency range of normal human heart rate is 0.67Hz-3Hz (corresponding to a heart rate of 40-180 times / minute), the passband of the filter is set to 0.5Hz-4Hz to ensure that the heart rate signal is fully retained while effectively suppressing out-of-band noise. The order of the filter is set to 4th order to avoid signal distortion while ensuring the filtering effect. The initial filtered signal obtained after bandpass filtering has obvious periodic characteristics.

[0041] Step 202: Calculate the signal-to-noise ratio of the initial filtered signal, and set the signal threshold according to the signal-to-noise ratio.

[0042] Specifically, in order to evaluate the quality of the initial filtered signal and determine the appropriate signal threshold, it is necessary to calculate the signal-to-noise ratio of the signal. In this embodiment, the power spectrum analysis method is used to calculate the signal-to-noise ratio. First, the initial filtered signal is subjected to a fast Fourier transform to obtain the spectrum of the signal; then the signal power PS is calculated within the main frequency band range (0.67Hz-3Hz) of the heart rate signal, and the noise power PN is calculated outside the frequency band range; the calculation formula of the signal-to-noise ratio SNR is: SNR = 10log10(PS / PN). In this embodiment, through a large amount of experimental data statistics, it is found that when the signal-to-noise ratio is greater than 15dB, the characteristics of the heart rate signal are more obvious and more suitable for identity recognition. The adaptive method is used to set the signal threshold, and the calculation formula of the signal threshold Th is: Th = α·RMS·10^(-SNR / 20), where RMS is the root mean square value of the initial filtered signal, and α is the adjustment coefficient. In this embodiment, α=1.5 is set. This threshold setting method takes into account the dual effects of signal strength and signal-to-noise ratio. When the signal-to-noise ratio is high, the threshold is relatively low, which can retain more valid signals; when the signal-to-noise ratio is low, the threshold is relatively high, which can effectively filter out noise interference.

[0043] Step 203: extracting signals greater than the signal threshold as valid signal segments.

[0044] Specifically, after the signal threshold is determined, the initial filtered signal is compared point by point, and continuous data segments with signal amplitudes greater than the signal threshold are extracted as valid signal segments.

[0045] Step 30: extract waveform characteristic parameters in the effective signal segment, and perform feature vector conversion on the waveform characteristic parameters according to a preset feature model to obtain a feature vector.

[0046] Specifically, each person's heart rate signal has unique waveform characteristics, which are reflected in the shape, amplitude distribution, and time characteristics of the waveform. In order to fully characterize the individual characteristics of the heart rate signal, it is necessary to extract multi-dimensional waveform feature parameters and convert these parameters into feature vectors that are easy for computers to process.

[0047] Based on the above embodiment, as an optional embodiment, the step of extracting waveform characteristic parameters in the effective signal segment may further include the following steps: Step 301: Calculate the waveform change rate in the effective signal segment, and identify the starting point and ending point of the dicrotic wave according to the waveform change rate.

[0048] Specifically, the embodiment of the present application uses the first-order difference method to calculate the waveform change rate in the effective signal segment. For the sampling sequence x(n), the calculation formula of the waveform change rate y(n) is: y(n)=[x(n+1)-x(n)] / T, where T is the sampling period. By calculating the waveform change rate, the fast-changing part of the signal can be highlighted, making the starting and ending features of the dicrotic wave more obvious. In order to suppress the influence of noise on the change rate calculation, the signal is smoothed by a 5-point moving average before the difference. After obtaining the waveform change rate, the present embodiment uses the threshold method to identify the starting point and the ending point of the dicrotic wave. Through statistical analysis, it is found that the starting point of the dicrotic wave usually corresponds to the positive zero-crossing point of the waveform change rate, and the subsequent change rate should exceed the preset threshold th1 (set to 20% of the signal root mean square value in this embodiment); the ending point of the dicrotic wave corresponds to the subsequent negative zero-crossing point, and the negative change rate before it should be less than the preset threshold th2 (set to -th1). In order to avoid false detection, it is also necessary to verify whether the time interval between adjacent starting points is within a reasonable range (set to 200ms-400ms in this embodiment).

[0049] Step 302: Calculate the area, peak value and duration of the dicrotic wave.

[0050] Specifically, after determining the starting point and the ending point of the dicrotic wave, this embodiment calculates three key characteristic parameters: dicrotic wave area (DWA), dicrotic wave peak (DWP) and dicrotic wave duration (DWD). Among them, DWA is obtained by numerically integrating the signal between the starting point and the ending point, and the trapezoidal integration method is used to improve the calculation accuracy; DWP is the maximum signal amplitude in the dicrotic wave interval, which is determined by searching for the local maximum in the interval; DWD is the time difference between the ending point and the starting point. DWA reflects the energy of the dicrotic wave and is related to the vascular rebound performance; DWP characterizes the intensity of the dicrotic wave and is related to the peripheral vascular resistance; DWD reflects the time characteristics of the dicrotic wave and is related to the vascular conduction velocity. Through this parameterized description, the morphological characteristics of the dicrotic wave can be converted into quantitative characteristic indicators.

[0051] Step 303: Calculate the dicrotic wave index according to the area, peak value and duration.

[0052] Specifically, this embodiment uses a weighted fusion method to calculate the dicrotic wave index DWI. The calculation formula is: DWI = w1·(DWA / DWAref)+w2·(DWP / DWPref)+w3·(DWD / DWDref), where DWA, DWP and DWD are the area, peak value and duration of the dicrotic wave, respectively, DWAref, DWPref and DWDref are the reference values ​​of the corresponding parameters (the mean obtained by a large number of sample statistics), w1, w2 and w3 are weight coefficients, and satisfy w1+w2+w3=1. In this embodiment, through the analysis of a large amount of experimental data, w1=0.4, w2=0.35, w3=0.25 are set. This set of weight configuration can better balance the contribution of the three parameters.

[0053] Step 304: extracting waveform characteristic parameters based on the dicrotic wave index.

[0054] Specifically, the mean MDWI and standard deviation SDDWI of the dicrotic wave index are calculated. These two parameters reflect the overall level and fluctuation degree of the dicrotic wave characteristics. Then the variation rate DWIVR between adjacent dicrotic wave indices is calculated, and its calculation formula is: DWIVR(n) = |DWI(n+1) - DWI(n)| / DWI(n), which reflects the temporal variation characteristics of the dicrotic wave characteristics. In order to capture the dynamic characteristics of the dicrotic wave index, the present embodiment also calculates the following parameters: the trend slope DWITS of the dicrotic wave index, which is obtained by linear regression of N consecutive DWI values; the fluctuation period DWIOP of the dicrotic wave index, which is determined by autocorrelation analysis to determine the main periodic components of the DWI sequence; the complexity DWIC of the dicrotic wave index, which uses the approximate entropy algorithm to calculate the complexity of the DWI sequence. These feature parameters extracted based on the dicrotic wave index form a multidimensional feature set, each of which reflects a certain aspect of the heart rate signal: MDWI and SDDWI describe the stability of the overall characteristics of the dicrotic wave, DWIVR reflects the short-term changes in the characteristics of the dicrotic wave, DWITS characterizes the long-term change trend of the characteristics of the dicrotic wave, DWIOP reflects the periodic law of the change of the dicrotic wave, and DWIC measures the complexity of the change of the dicrotic wave.

[0055] Based on the above embodiment, as an optional embodiment, the step of converting the waveform characteristic parameters into characteristic vectors according to the preset characteristic model to obtain the characteristic vectors may also include the following steps: Step 401: Input waveform feature parameters into a preset feature model to obtain a feature correlation matrix.

[0056] Specifically, since there may be correlation between the extracted waveform feature parameters, directly using these features may lead to information redundancy and affect the recognition effect. Therefore, it is necessary to analyze the correlation between the features, optimize the feature combination by clustering and weighting, and finally obtain efficient feature representation by dimensionality reduction. First, the waveform feature parameters are input into the preset feature model to calculate the feature correlation matrix. The preset feature model adopts a mixed Gaussian model built based on large-scale training data. For N feature parameters, the mutual information between them is calculated to form an N×N feature correlation matrix R. The matrix element R(i,j) represents the degree of correlation between the i-th feature and the j-th feature, and the value range is [0,1]. The larger the value, the stronger the correlation.

[0057] Step 402: Perform feature clustering analysis based on the feature correlation matrix, and calculate the weight coefficient of each feature according to the clustering analysis results.

[0058] Specifically, based on the feature correlation matrix, this embodiment uses a spectral clustering algorithm to perform feature clustering analysis. First, construct the Laplace matrix L=DR of the feature, where D is the degree matrix (diagonal matrix, the diagonal elements are the sum of each row of the correlation matrix); then calculate the eigenvalues ​​and eigenvectors of L, select the eigenvectors corresponding to the smallest k eigenvalues ​​to form a transformation matrix; finally, use the K-means algorithm to cluster the transformed features. Through experimental analysis, this embodiment sets the number of clusters k to 3, and divides the features into high correlation groups, medium correlation groups, and low correlation groups.

[0059] Step 403: Perform weighted combination on each feature based on the weight coefficient.

[0060] Specifically, for each group of features obtained by clustering, its weight coefficient is calculated. The weight calculation takes into account two factors: intra-group correlation IC and inter-group discrimination ID. For the i-th group of features, the calculation formula of its weight coefficient wi is: wi = (1-ICi)·IDi / ∑[(1-ICj)·IDj], where ICi is the average correlation of the features in the i-th group, and IDi is the average distance between the i-th group and other groups. This weight distribution method takes into account both the independence of the features (low internal correlation) and the ability to distinguish the features (high inter-group difference). After obtaining the weight coefficients of each group of features, the features in each group are weighted combined. Suppose the i-th group contains mi features, denoted as fi1, fi2, ..., fimi, then the calculation formula of the weighted combination result Fi of the group is: Fi = wi·∑(αij·fij), where αij is the normalization coefficient of the features in the group to ensure that features of different dimensions are comparable. In this way, the original N features are compressed into three combined features F1, F2 and F3.

[0061] Step 404: Perform dimensionality reduction processing on the weighted combination result to obtain a feature vector.

[0062] Specifically, in order to further optimize the feature representation, this embodiment uses a local linear embedding algorithm to reduce the dimension of the weighted combination result. The algorithm can maintain the local structural characteristics of the data and is suitable for processing nonlinear data. Through dimensionality reduction, the 3D combined features are mapped to a 2D feature space to obtain the final feature vector V=[v1,v2].

[0063] Step 40: Match the feature vector with the preset authorized user template to obtain the identity authentication result of the target person.

[0064] Specifically, in this embodiment, in order to accurately determine whether the current user is an authorized user, it is necessary to match and compare the extracted feature vector with the pre-stored authorized user template. Since the heart rate signal has a certain time-varying property, simply using a hard matching method with a fixed threshold may lead to a high misjudgment rate. Therefore, this embodiment adopts an adaptive multi-level matching strategy to improve the accuracy and reliability of identity authentication.

[0065] Based on the above embodiment, as an optional embodiment, the step of matching the feature vector with a preset authorized user template to obtain the identity authentication result of the target person may further include the following steps: Step 401: Calculate feature difference values ​​between the feature vector and each template feature vector in a preset authorized user template.

[0066] Specifically, in this embodiment, considering that the heart rate signal is greatly affected by the human body state and environmental factors, the matching of a single feature vector may be unstable. Therefore, multiple feature vectors are stored in the preset authorized user template to form a template feature vector set. At the same time, a dynamic threshold mechanism is adopted to adapt to the time-varying characteristics of feature matching and improve the adaptability and reliability of verification. Calculate the feature difference value between the current feature vector and each vector in the template feature vector set. Suppose the current feature vector is V = [v1, v2], and the template feature vector set is T = {T1, T2, ..., TM}, where Ti = [ti1, ti2] represents the i-th template feature vector, and the feature difference value Di is calculated using the weighted Euclidean distance to obtain the feature difference value sequence {D1, D2, ..., DM}.

[0067] Step 402: construct a similarity score based on the feature difference values.

[0068] Specifically, in order to comprehensively consider the matching of multiple template feature vectors, a weighted average method based on the kernel function is used to calculate the similarity score. The specific formula is: S=∑[K(Di)] / M, where K(x) is the Gaussian kernel function K(x)=exp(-x² / σ²), σ is the kernel function bandwidth parameter, and the optimal value is determined by cross-validation. This scoring method can reasonably smooth the impact of feature differences and provide a more stable similarity measurement.

[0069] Step 403: Setting a dynamic matching threshold based on historical matching data.

[0070] Specifically, based on the historical matching data, the similarity scores of the most recent N (N=50 in this embodiment) successful verifications are statistically analyzed to calculate their mean μ and standard deviation σ. The calculation formula for the dynamic matching threshold Th is: Th =μ-k·σ, where k is the adjustment coefficient (k=2.5 in this embodiment). This dynamic threshold can adaptively track the changing trend of feature matching, improving the adaptability of the system while maintaining the reliability of verification.

[0071] Step 404: When the similarity score is greater than the dynamic matching threshold, it is confirmed that the identity authentication is passed.

[0072] Step 405: When the similarity score is not greater than the dynamic matching threshold, it is confirmed that the identity authentication fails.

[0073] Specifically, when the calculated similarity score S is greater than the dynamic matching threshold Th, the current user identity authentication is confirmed to be passed, and the subsequent unlocking operation is allowed. When S is not greater than Th, the identity authentication is confirmed to have failed, and the system remains locked. In order to prevent continuous failed attempts, the system also sets a verification interval control mechanism, that is, after three consecutive verification failures, it is necessary to wait 30 seconds before the next verification attempt.

[0074] See also Figure 2 , is a module diagram of an identity recognition system based on a pressure sensor provided in an embodiment of the present application. The identity recognition system based on a pressure sensor may include: a data acquisition module, a data processing module, a data conversion module and a data verification module, wherein: A data acquisition module, used to acquire a pressure signal of a target person collected by a pressure sensor and convert the pressure signal into heart rate data; A data processing module, used for performing data preprocessing on the heart rate data to obtain a valid signal segment; A data conversion module, used for extracting waveform characteristic parameters in the effective signal segment, and performing characteristic vector conversion on the waveform characteristic parameters according to a preset characteristic model to obtain a characteristic vector; The data verification module is used to match the feature vector with a preset authorized user template to obtain the identity verification result of the target person.

[0075] Optionally, the data acquisition module is further used to charge amplify the pressure signal to obtain an amplified voltage signal; Performing analog-to-digital conversion on the amplified voltage signal to obtain a digital quantity signal; Calculating peak values ​​and trough values ​​according to the digital signal, and determining a heart rate cycle based on the peak values ​​and trough values; Heart rate data is generated according to the heart rate cycle.

[0076] Optionally, the data processing module is further used to perform bandpass filtering on the heart rate data to obtain an initial filtered signal; Calculating a signal-to-noise ratio of the initial filtered signal, and setting a signal threshold according to the signal-to-noise ratio; The signal that is greater than the signal threshold is extracted as a valid signal segment.

[0077] Optionally, the data conversion module is further used to calculate the waveform change rate in the effective signal segment, and identify the starting point and the ending point of the dicrotic wave according to the waveform change rate; Calculate the area, peak value and duration of the dicrotic wave; Calculating a dicrotic wave index based on the area, peak value and duration; Waveform characteristic parameters are extracted based on the dicrotic wave index.

[0078] Optionally, the data conversion module is further used to input the waveform characteristic parameters into a preset characteristic model to obtain a characteristic correlation matrix; Performing feature clustering analysis based on the feature correlation matrix, and calculating weight coefficients of various features according to the clustering analysis results; Performing a weighted combination of the features based on the weight coefficients; The weighted combination result is subjected to dimensionality reduction processing to obtain a feature vector.

[0079] Optionally, the data verification module is further used to calculate a feature difference value between the feature vector and each template feature vector in a preset authorized user template; constructing a similarity score based on the feature difference values; Set dynamic matching thresholds based on historical matching data; When the similarity score is greater than the dynamic matching threshold, confirming that the identity authentication is passed; When the similarity score is not greater than the dynamic matching threshold, it is determined that the identity authentication has failed.

[0080] Optionally, the data acquisition module is further used to monitor the amplitude change rate of the pressure signal in real time, and when the amplitude change rate exceeds a preset range, it is determined to be an abnormal fluctuation; Calculating the duration and amplitude of the abnormal fluctuation; When the duration is less than the time threshold and the fluctuation amplitude is less than the amplitude threshold, smoothing the abnormal fluctuation; When the duration is not less than the time threshold or the fluctuation amplitude is not less than the amplitude threshold, the currently collected pressure signal is marked as an invalid signal, and the pressure signal of the target person collected by the pressure sensor is reacquired until a valid pressure signal is collected.

[0081] It should be noted that: when the system provided in the above embodiment realizes its functions, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0082] An embodiment of the present application also provides a computer storage medium, which can store multiple instructions. The instructions are suitable for being loaded by a processor and executing a pressure sensor-based identity recognition method of the above embodiment. The specific execution process can be found in the specific description of the above embodiment, which will not be repeated here.

[0083] Please refer to Figure 3 The application also discloses an electronic device. Figure 3 The electronic device 300 may include: at least one processor 301 , at least one network interface 304 , a user interface 303 , a memory 305 , and at least one communication bus 302 .

[0084] The communication bus 302 is used to realize the connection and communication between these components.

[0085] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0086] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0087] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0088] Among them, the memory 305 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may optionally also be at least one storage device located away from the aforementioned processor 301. Refer to Figure 3 , the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of an identity recognition method based on a pressure sensor.

[0089] exist Figure 3In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 301 can be used to call the application program stored in the memory 305 for a pressure sensor-based identity recognition method. When executed by one or more processors 301, the electronic device 300 executes one or more of the methods described in the above embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited to the described order of actions, because according to the present application, 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 the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0090] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0091] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, 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 service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0092] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0093] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0094] 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 computer-readable memory. Based on this understanding, the technical solution of the present application, 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, which is stored in a memory and includes several instructions for 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 application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0095] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0096] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. An identity recognition method based on a pressure sensor, characterized in that: The method comprises: Acquire a pressure signal of a target person collected by a pressure sensor, and convert the pressure signal into heart rate data; Performing data preprocessing on the heart rate data to obtain a valid signal segment; Extracting waveform characteristic parameters from the effective signal segment, and performing feature vector conversion on the waveform characteristic parameters according to a preset feature model to obtain a feature vector; The feature vector is matched with a preset authorized user template to obtain the identity authentication result of the target person.

2. The identity recognition method based on pressure sensor according to claim 1, characterized in that: The step of converting the pressure signal into heart rate data comprises: Performing charge amplification on the pressure signal to obtain an amplified voltage signal; Performing analog-to-digital conversion on the amplified voltage signal to obtain a digital quantity signal; Calculating peak values ​​and trough values ​​according to the digital signal, and determining a heart rate cycle based on the peak values ​​and trough values; Heart rate data is generated according to the heart rate cycle.

3. The identity recognition method based on pressure sensor according to claim 1, characterized in that: The step of performing data preprocessing on the heart rate data to obtain a valid signal segment includes: Performing bandpass filtering on the heart rate data to obtain an initial filtered signal; Calculating a signal-to-noise ratio of the initial filtered signal, and setting a signal threshold according to the signal-to-noise ratio; The signal that is greater than the signal threshold is extracted as a valid signal segment.

4. The identity recognition method based on pressure sensor according to claim 1, characterized in that: The step of extracting waveform characteristic parameters from the effective signal segment comprises: Calculating the waveform change rate in the effective signal segment, and identifying the starting point and the ending point of the dicrotic wave according to the waveform change rate; Calculate the area, peak value and duration of the dicrotic wave; Calculating a dicrotic wave index based on the area, peak value and duration; Waveform characteristic parameters are extracted based on the dicrotic wave index.

5. The identity recognition method based on pressure sensor according to claim 1, characterized in that: The step of converting the waveform characteristic parameters into characteristic vectors according to a preset characteristic model to obtain characteristic vectors includes: Inputting the waveform characteristic parameters into a preset characteristic model to obtain a characteristic correlation matrix; Performing feature clustering analysis based on the feature correlation matrix, and calculating weight coefficients of various features according to the clustering analysis results; Performing a weighted combination of the features based on the weight coefficients; The weighted combination result is subjected to dimensionality reduction processing to obtain the feature vector.

6. The identity recognition method based on pressure sensor according to claim 1, characterized in that: The matching of the feature vector with a preset authorized user template to obtain the identity authentication result of the target person includes: Calculate a feature difference value between the feature vector and each template feature vector in a preset authorized user template; constructing a similarity score based on the feature difference values; Set dynamic matching thresholds based on historical matching data; When the similarity score is greater than the dynamic matching threshold, confirming that the identity authentication is passed; When the similarity score is not greater than the dynamic matching threshold, it is determined that the identity authentication has failed.

7. The identity recognition method based on pressure sensor according to claim 1, characterized in that: After obtaining the pressure signal of the target person collected by the pressure sensor, the method further includes: monitoring the amplitude change rate of the pressure signal in real time, and determining it as an abnormal fluctuation when the amplitude change rate exceeds a preset range; Calculating the duration and amplitude of the abnormal fluctuation; When the duration is less than the time threshold and the fluctuation amplitude is less than the amplitude threshold, smoothing the abnormal fluctuation; When the duration is not less than the time threshold or the fluctuation amplitude is not less than the amplitude threshold, the currently collected pressure signal is marked as an invalid signal, and the pressure signal of the target person collected by the pressure sensor is reacquired until a valid pressure signal is collected.

8. An identity recognition system based on a pressure sensor, characterized in that: The system comprises: A data acquisition module, used to acquire a pressure signal of a target person collected by a pressure sensor and convert the pressure signal into heart rate data; A data processing module, used for performing data preprocessing on the heart rate data to obtain a valid signal segment; A data conversion module, used for extracting waveform characteristic parameters in the effective signal segment, and performing characteristic vector conversion on the waveform characteristic parameters according to a preset characteristic model to obtain a characteristic vector; The data verification module is used to match the feature vector with a preset authorized user template to obtain the identity verification result of the target person.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-7.

Images