Living body detection method, medium, product, equipment, system and vehicle

Abstract

The invention relates to a living body detection method, a medium, a product, equipment, a system and a vehicle. The living body detection method comprises the steps of determining whether a living body exists in a vehicle or not according to a response signal matrix; wherein the response signal matrix comprises multipath echo signals of the detection signals. According to the invention, in-vehicle living body detection is realized, and the accuracy of living body detection is improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a liveness detection method, medium, product, equipment, system and vehicle. Background Art

[0002] After parking, users may accidentally leave children or pets inside their vehicles, whether due to inadvertent or temporary reasons. In a closed vehicle, temperature and air quality can quickly reach critical levels, posing a serious threat to the safety of living animals. Implementing liveness detection to improve safety is a pressing technical challenge. Summary of the Invention

[0003] The embodiments of the present application provide a liveness detection method, medium, product, device, system, and vehicle, which implement in-vehicle liveness detection and improve the accuracy of liveness detection, thereby at least partially solving the above-mentioned technical problems.

[0004] To achieve the above-mentioned purpose, according to a first aspect of the present application, a living body detection method is provided, comprising: determining whether a living body exists in a vehicle based on a response signal matrix; wherein the response signal matrix includes multipath echo signals of a detection signal.

[0005] Optionally, the detection signal is an ultra-wideband signal.

[0006] Optionally, the signals in the response signal matrix are channel impulse response signals.

[0007] Optionally, determining whether there is a living body in the vehicle according to the response signal matrix includes: determining whether there is a moving target in the vehicle according to the response signal matrix; wherein the living body includes the moving target.

[0008] Optionally, determining whether there is a moving target in the vehicle based on the response signal matrix includes: determining a response signal standard deviation based on the response signal matrix; and determining whether there is a moving target in the vehicle based on the response signal standard deviation.

[0009] Optionally, determining the response signal standard deviation according to the response signal matrix includes: selecting a first signal from the response signal matrix; and determining the response signal standard deviation according to the first signal.

[0010] Optionally, the first signal is an echo signal corresponding to a vehicle door.

[0011] Optionally, determining the response signal standard deviation according to the response signal matrix includes: determining the standard deviation of multiple signals in the response signal matrix; and determining the response signal standard deviation according to the standard deviations of the multiple signals.

[0012] Optionally, determining the standard deviation of the response signal based on the standard deviations of the plurality of signals includes: averaging the standard deviations of the plurality of signals to obtain the standard deviation of the response signal.

[0013] Optionally, determining whether there is a moving target in the vehicle based on the standard deviation of the response signal includes: determining whether there is a moving target in the vehicle based on a comparison between the standard deviation of the response signal and a first threshold.

[0014] Optionally, determining whether there is a living body in the vehicle according to the response signal matrix includes: determining whether there is a static target in the vehicle according to the response signal matrix; wherein the living body includes the static target.

[0015] Optionally, determining whether there is a stationary target in the vehicle according to the response signal matrix includes: determining a second signal according to the response signal matrix; and determining whether there is a stationary target in the vehicle according to the second signal.

[0016] Optionally, determining the second signal according to the response signal matrix includes: performing enhancement processing on the signal in the response signal matrix to obtain the enhanced response signal matrix; and determining the second signal according to the enhanced response signal matrix.

[0017] Optionally, the performing enhancement processing on the signals in the response signal matrix to obtain the enhanced response signal matrix includes: performing band-pass filtering processing on the signals in the response signal matrix to obtain the enhanced response signal matrix.

[0018] Optionally, the performing enhancement processing on the signals in the response signal matrix to obtain the enhanced response signal matrix includes: performing projection processing on the signals in the response signal matrix to obtain the enhanced response signal matrix.

[0019] Optionally, the projection processing is performed on the signals in the response signal matrix to obtain the enhanced response signal matrix, including: for each signal in the response signal matrix, obtaining the projection components of the signal on multiple projection planes; selecting the projected signal from the multiple projection components; and generating the enhanced response signal matrix based on the multiple projected signals.

[0020] Optionally, performing enhancement processing on the signals in the response signal matrix to obtain the enhanced response signal matrix includes: performing signal expansion processing on the response signal matrix to obtain the enhanced response signal matrix.

[0021] Optionally, determining the second signal according to the enhanced response signal matrix includes: selecting the second signal from the enhanced response signal matrix; wherein the second signal is a signal with the largest standard deviation in the enhanced response signal matrix.

[0022] Optionally, determining whether there is a stationary target in the vehicle according to the second signal includes: determining a target frequency of the second signal; and determining whether there is a stationary target in the vehicle according to the target frequency.

[0023] Optionally, determining the target frequency of the second signal includes: performing spectrum conversion processing on the second signal to obtain multiple frequencies; and determining the target frequency according to the multiple frequencies.

[0024] Optionally, determining the target frequency based on the multiple frequencies includes: selecting a target frequency from the multiple frequencies; wherein the target frequency is a frequency with the largest modulus value among the multiple frequencies.

[0025] Optionally, determining whether there is a stationary target in the vehicle according to the target frequency includes: determining whether there is a stationary target in the vehicle according to a comparison between the target frequency and a second threshold.

[0026] Optionally, determining whether there is a stationary target in the vehicle based on the second signal includes: determining an autocorrelation function of the second signal; and determining whether there is a stationary target in the vehicle based on the autocorrelation function.

[0027] Optionally, determining whether there is a stationary target in the vehicle according to the autocorrelation function includes: determining whether there is a stationary target in the vehicle according to a comparison between a slope of the autocorrelation function and a third threshold.

[0028] Optionally, determining whether there is a static target in the vehicle based on the second signal includes: determining a primary frequency modulus and a secondary frequency modulus of the second signal; and determining whether there is a static target in the vehicle based on the primary frequency modulus and the secondary frequency modulus.

[0029] Optionally, determining whether there is a static target in the vehicle based on the main frequency modulus and the secondary frequency modulus includes: determining whether there is a static target in the vehicle based on comparing the sum of the ratios of the secondary frequency modulus to the main frequency modulus with a fourth threshold.

[0030] Optionally, the method further includes: performing data preprocessing on the response signal matrix to obtain the preprocessed response signal matrix; wherein the preprocessed response signal matrix is used to determine whether there is a living body in the vehicle.

[0031] Optionally, performing data preprocessing on the response signal matrix to obtain the preprocessed response signal matrix includes: performing outlier filtering on the response signal matrix to obtain the preprocessed response signal matrix.

[0032] Optionally, performing outlier filtering on the response signal matrix to obtain the preprocessed response signal matrix includes: for each signal in the response signal matrix, determining the preprocessed signal based on a comparison of the signal with the median absolute deviation of the window in which the signal is located; wherein the preprocessed response signal matrix includes multiple preprocessed signals.

[0033] Optionally, performing data preprocessing on the response signal matrix to obtain the preprocessed response signal matrix includes: performing centralization processing on the response signal matrix to obtain the preprocessed response signal matrix.

[0034] Optionally, the centralizing the response signal matrix to obtain the preprocessed response signal matrix includes: determining a signal mean in the response signal matrix; and subtracting each signal in the response signal matrix from the signal mean to obtain the preprocessed response signal matrix.

[0035] According to a second aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned living body detection method is implemented.

[0036] According to a third aspect of the present application, a computer program product is provided, comprising a computer program, wherein the computer program implements the above-mentioned liveness detection method when executed by a processor.

[0037] According to a fourth aspect of the present application, an electronic device is provided, comprising: a memory storing a computer program; and a processor configured to execute the computer program in the memory to implement the above-mentioned liveness detection method.

[0038] According to a fifth aspect of the present application, a liveness detection system is provided, which includes a first radar and an electronic device; wherein the first radar is used to: transmit a detection signal and receive a multipath echo signal of the detection signal; the electronic device is used to: determine whether there is a living body in the vehicle based on a response signal matrix; wherein the response signal matrix includes the multipath echo signal.

[0039] Optionally, the first radar is of self-transmitting and self-receiving type.

[0040] Optionally, the installation position of the first radar includes any one of the following: the center of the roof, the center of the passenger compartment.

[0041] Optionally, the first radar includes two radars, and the type of the first radar is one receiving and one transmitting.

[0042] Optionally, the installation position of the first radar includes any one of the following: the center position of the front side of the roof and the center position of the rear side of the roof, the center position of the left side of the roof and the center position of the right side of the roof, the center position of the front side of the passenger compartment and the center position of the rear side of the passenger compartment, the center position of the left side of the passenger compartment and the center position of the right side of the passenger compartment.

[0043] According to a sixth aspect of the present application, a vehicle is provided, comprising the above-mentioned electronic device or the above-mentioned living body detection system.

[0044] The technical solution provided by the embodiment of the present application determines whether a living being is present in a vehicle based on a response signal matrix. The response signal matrix includes multipath echo signals of the detection signal. Since the echo signal is formed by the reflection of the detection signal, and the echo signals reflected by different objects or living beings, as well as different states of living beings, differ, such as signal frequency or signal standard deviation, the embodiment of the present application can analyze the parameters of the echo signal based on the response signal matrix to achieve in-vehicle living being detection. Furthermore, the embodiment of the present application can effectively solve the problem of signal superposition through the method of virtual multipath, which helps to extract more accurate living being features, thereby improving the accuracy of living being detection.

[0045] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] To more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present application. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0047] In order to more completely understand the present application and its beneficial effects, the following description will be given in conjunction with the accompanying drawings, wherein the same drawing numbers represent the same parts in the following description.

[0048] Figure 1 This is a flow chart of a liveness detection method provided in an embodiment of the present application;

[0049] Figure 2 is a schematic diagram of a vehicle provided in an embodiment of the present application;

[0050] Figure 3 is a schematic diagram of a radar coverage area provided in an embodiment of the present application;

[0051] Figure 4 is a schematic diagram of a radar installation position provided in an embodiment of the present application;

[0052] Figure 5This is a flowchart of another liveness detection method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the embodiments described are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0054] According to the first aspect of the present application, an embodiment of the present application provides a liveness detection method.

[0055] See also Figure 1 , Figure 1 This is a flow chart of a liveness detection method provided in an embodiment of the present application. Figure 1 As shown, the living body detection method may include the following steps:

[0056] Step S100: determining whether there is a living body in the vehicle according to a response signal matrix; wherein the response signal matrix includes multipath echo signals of the detection signal.

[0057] The detection signal is a signal emitted by the vehicle for detecting living things. The embodiment of the present application does not limit the specific type of the detection signal. Optionally, the detection signal can be a radio wave signal (such as a millimeter wave signal, an ultra-wideband signal), a laser signal, or an invisible light signal (such as an infrared signal). In some embodiments, the detection signal can be an ultra-wideband (Ultra-WideBand, UWB) signal, so as to reduce cost and power consumption. In some embodiments, the detection signal can be emitted once or multiple times. For example, the detection signal can be emitted periodically. In some embodiments, the detection signal can be emitted when the living thing detection condition is met. Among them, the living thing detection condition can be flexibly set according to actual needs, and the embodiment of the present application does not limit this. For example, the living thing detection condition includes the door being closed, such as the door being closed but not locked and the door being closed and locked.

[0058] After contacting a target such as an object or living organism, the detection signal is reflected, forming an echo signal. However, due to the presence of various objects or living organisms in and within the vehicle, such as seats, armchairs, and doors, the detection signal is reflected multiple times, forming a multipath echo signal. A response signal matrix can be generated based on the multipath echo signals. In some embodiments, the echo signal of the detection signal can be a channel impulse response (CIR) signal, so that the signal in the response signal matrix is a channel impulse response signal, and the response signal matrix can also be referred to as a CIR matrix.

[0059] The response signal matrix may include two dimensions: fast time and slow time. The fast time dimension is the time domain pulse position of the signal, which represents the echo signal information at different distances in space and is used to determine the distance information of the target. The slow time dimension is the change between continuous pulses of the signal. By analyzing the phase change between pulses, the speed or motion state of the target can be determined. In some embodiments, each signal in the response signal matrix may include: Tap number, cycle number, and corresponding I (In-phase, in-phase component) / Q (Quadrature, orthogonal component) complex plane data. Among them, each Tap refers to a major multipath component, corresponding to the arrival time of the transmitted detection signal after direct or multiple reflections, which can provide high-precision distance information. In practical applications, the Tap can be equivalent to a sampling point at a given distance.

[0060] In an embodiment of the present application, after the detection signal is transmitted, the first received signal may be a direct wave formed by the direct reception of the detection signal, rather than an echo signal formed after the detection signal is reflected, so the first one or more signals received can be eliminated. In some embodiments, a signal with Tap number [a, b] can be used to generate a response signal matrix, that is, within a single transmission cycle of the detection signal, the signal with Tap number a to b is selected to generate a response signal matrix. The response signal matrix consists of two dimensions: fast time and slow time. The fast time dimension corresponds to the propagation time of the signal in the distance dimension. By sampling the echo signal of a single pulse, the distance-related information of the target can be obtained; the slow time dimension involves the repetition interval of the pulse sequence. The echo received after the pulse is transmitted and the same time delay is related to the speed and azimuth information of the target. Thus, the response signal matrix can be shown by the following formula 1.

[0061] Formula 1: R = [r1 r2 … r m ]

[0062] Among them, R is the response signal matrix; r m is the signal with Tap number m, which can be expressed by the following formula 2.

[0063] Formula 2: r m =I m +Q m ×j

[0064] Among them, I m is the signal r m The real part, Q m is the signal r m The imaginary part of . Through the response signal matrix, the relevant reflection information in the time and space dimensions corresponding to the current sensing area can be obtained.

[0065] Based on the response signal matrix, it can be determined whether there is a living body in the car. For example, based on the response signal matrix, the standard deviation, frequency and / or signal-to-noise ratio of the signal can be analyzed to determine whether there is a living body in the car. For an introduction to determining whether there is a living body based on the response signal matrix, please refer to the following embodiments, which will not be elaborated here. A living body refers to an organism with life characteristics, including humans and pets. In the embodiments of the present application, living bodies can be divided into moving targets and static targets. A moving target can also be called a dynamic living body, which refers to a living body that shows obvious signs of life activities, such as the movement of a person or pet. A static target can also be called a static living body, which refers to a living body that does not show obvious signs of life activities but still has life characteristics. For example, a motionless person or pet, although there is no obvious movement, still has weak vital signs such as breathing and heartbeat.

[0066] In some embodiments, if a living being is determined to be present in the vehicle, such as a moving and / or stationary object, a target operation can be performed to alert the vehicle owner, driver, or people around the vehicle of the presence of a living being. The present application does not limit the specific content of the target operation, and in actual application, the target operation can be flexibly configured based on actual needs. For example, the target operation includes but is not limited to: opening a window, turning on the air conditioner, sounding an alarm, sending a text message, voice broadcasting, turning on the alarm, etc.

[0067] In summary, the technical solution provided by the embodiments of the present application determines whether a living being is present in a vehicle based on a response signal matrix. The response signal matrix includes multipath echo signals of the detection signal. Since the echo signal is formed by the reflection of the detection signal, and the echo signals reflected by different objects or living beings, as well as different states of living beings, differ, such as signal frequency or signal standard deviation, the embodiments of the present application can analyze the parameters of the echo signal based on the response signal matrix to achieve in-vehicle living being detection. Furthermore, the embodiments of the present application can effectively solve the problem of signal superposition through the use of a virtual multipath method, which helps to extract more accurate living being features, thereby improving the accuracy of living being detection.

[0068] In some embodiments, the above method may include the following steps:

[0069] Step S010: performing data preprocessing on the response signal matrix to obtain a preprocessed response signal matrix.

[0070] The preprocessed response signal matrix is used to determine whether a living organism is present in the vehicle. That is, the response signal matrix in step S100 can be the preprocessed response signal matrix obtained in step S010. Data preprocessing can improve the accuracy of the signals in the response signal matrix, facilitating subsequent accurate and rapid liveness detection. The present embodiment of the application does not limit the specific method of data preprocessing; in actual applications, data preprocessing can be flexibly configured based on actual needs. For example, data preprocessing includes, but is not limited to, outlier filtering, data zero-meaning, and other processing.

[0071] In some embodiments, the above step S010 may include the following steps:

[0072] Step S011: performing outlier filtering on the response signal matrix to obtain a pre-processed response signal matrix.

[0073] Outlier filtering is used to remove abnormal echo signals. Due to human motion, environmental factors, and equipment noise, occasional outliers may occur. The outlier filtering implemented in this embodiment of the application can eliminate abnormal outliers while preserving the overall distribution of the response signal matrix.

[0074] Among them, outlier filtering can be based on the Hampel filter, which can filter from two dimensions: I signal and Q signal. The Hampel filter is a linear filter based on the median outlier detection method. m , for the data X that needs to be filtered s , select a fixed-length window and use the median absolute deviation to determine X s If the data X s If it is judged to be valid, the output is X s ; Otherwise, use the window value instead of X s Based on this, in some embodiments, step S011 may include: for each signal in the response signal matrix, determining a preprocessed signal based on a comparison of the signal with the median absolute deviation of the window in which the signal resides. The preprocessed response signal matrix includes multiple preprocessed signals.

[0075] In some embodiments, the above step S010 may include the following steps:

[0076] Step S012: performing centralization processing on the response signal matrix to obtain a pre-processed response signal matrix.

[0077] Centralization can also be called zero-mean processing. The core is to subtract the mean of the data set from each data point. In this way, the average value of the processed data in each dimension will be zero, eliminating the "DC component" of the data, that is, the average value of the data.

[0078] In the embodiments of the present application, the amplitudes of echo signals from different tap numbers are all different. Centralization is used to perform a translation transformation on the multipath echo signals, facilitating subsequent liveness detection. Based on this, in some embodiments, step S012 may include: determining a signal mean in the response signal matrix; and subtracting the signal mean from each signal in the response signal matrix to obtain a preprocessed response signal matrix.

[0079] In practical applications, outlier filtering and centralization processing can be used in combination. For example, centralization processing is used after outlier filtering to obtain a preprocessed response signal matrix; or, outlier filtering is used after centralization processing to obtain a preprocessed response signal matrix. Of course, other data preprocessing methods can also be used in combination, and the embodiments of the present application are not limited to this. Exemplarily, the above-mentioned step S010 may include: performing outlier filtering on the response signal matrix to obtain a response signal matrix after outlier filtering; performing centralization processing on the response signal matrix after outlier filtering to obtain a centralized response signal matrix. Among them, the centralized response signal matrix, that is, the preprocessed response signal matrix, can be used for liveness detection.

[0080] With signal r m For example, the signal r′ after outlier filtering m It can be shown as the following formula 3, so the response signal matrix after outlier filtering can be shown as the following formula 4.

[0081] Formula 3: r′ m =I′ m +Q′ m ×j

[0082] Formula 4: R′=[r′1 r′2 … r′ m ]

[0083] Among them, I′ m is the signal r′ m The real part of Q′ can be calculated by the following formula 5; n is the signal r′ m The imaginary part of can be calculated by the following formula 6.

[0084] Formula 5: I′ m =hampel(I m , k, nsigma)

[0085] Formula 6: Q′ m =hampel(Q m , k, nsigma)

[0086] Among them, k is the sample point X sThe number of neighbors, so that the window size of the hampel filter can be 2×k+1; nsigma is the median absolute deviation value.

[0087] After filtering outliers, the signal is centered and the signal r″ m It can be shown as the following formula 7, so the centralized response signal matrix can be shown as the following formula 8.

[0088] Formula 7: r" m =I″ m +Q″ m ×j

[0089] Formula 8: R"=[r"1 r"2 ... r" m ]

[0090] Among them, I m is the signal r″ m The real part of Q can be calculated by the following formula 9: n is the signal r″ m The imaginary part of can be calculated using the following formula 10.

[0091] Formula 9: I″ m =I′ m -mean(I′ m )

[0092] Formula 10: Q″ m =Q′-mean(Q′)

[0093] Among them, mean is the average processing.

[0094] In summary, the technical solution provided by the embodiments of this application preprocesses the acquired response signal matrix and then determines whether a living person is present in the vehicle based on the preprocessed response signal matrix. This data preprocessing improves the accuracy of the signals in the response signal matrix, facilitating subsequent accurate and rapid liveness detection.

[0095] In some embodiments, the above step S100 may include the following steps:

[0096] Step S110: determining whether there is a moving target in the vehicle based on the response signal matrix;

[0097] In some embodiments, the above step S100 may include the following steps:

[0098] Step S120: Determine whether there is a static target in the vehicle according to the response signal matrix.

[0099] Among them, living objects include moving targets and static targets. Moving targets, also known as dynamic living objects, refer to living objects that show obvious signs of life activities, such as the movement of people or pets. Static targets, also known as static living objects, refer to living objects that do not show obvious signs of life activities but still have life characteristics. For example, a motionless person or pet, although without obvious movements, still has weak vital signs such as breathing and heartbeat. The embodiment of the present application divides living objects into moving targets and static targets, which can solve the problem of signal leakage outside the vehicle.

[0100] Since moving targets have more obvious signs of life activity than static targets, they are easier to detect. Therefore, the embodiment of the present application can first determine whether there is a moving target in the car based on the response signal matrix. If there is no moving target in the car, it can then determine whether there is a static target in the car based on the response signal matrix. Among them, if there is a moving target in the car, the liveness detection result can be directly determined, such as performing operations such as opening the window, turning on the air conditioner or alarming, without performing static target detection. Of course, in actual applications, if there is a moving target in the car, static targets can also be further detected, and the embodiment of the present application does not limit this. In addition, in actual applications, the detection of moving targets and static targets can also be performed simultaneously, that is, the execution of the above steps S110 and S120 does not affect each other, and the embodiment of the present application does not limit this.

[0101] In some embodiments, the above step S110 may include the following steps:

[0102] Step S111: determining the response signal standard deviation according to the response signal matrix;

[0103] Step S112: Determine whether there is a moving target in the vehicle based on the standard deviation of the response signal.

[0104] In a normal vehicle interior, if only noise is present, the response signal standard deviation will be relatively weak. However, if an adult or child is moving around in the vehicle, the response signal standard deviation will be larger due to the superposition of multipath effects, exceeding normal fluctuations. In embodiments of the present application, the response signal standard deviation is determined based on a response signal matrix, such as a preprocessed response signal matrix, and then the presence of a moving target in the vehicle is determined based on the response signal standard deviation. In some embodiments, step S112 may include determining the presence of a moving target in the vehicle based on a comparison of the response signal standard deviation and a first threshold. If the response signal standard deviation is greater than the first threshold, a moving target is determined to be present in the vehicle; if the response signal standard deviation is less than the first threshold, a moving target is determined to be absent. If the response signal standard deviation is equal to the first threshold, it can be assumed that a moving target is present or absent in the vehicle. This can be flexibly set based on actual needs. Furthermore, the first threshold can be a preset threshold, for example, determined based on experimental testing of living body movement or based on experience. This can be flexibly set based on actual needs.

[0105] In some embodiments, step S111 may include: selecting a first signal from a response signal matrix; and determining a standard deviation of the response signal based on the first signal. The first signal may be any one of the signals in the response signal matrix; or, the first signal may be a signal in the response signal matrix that satisfies certain conditions. For example, the first signal may be an echo signal corresponding to a vehicle door, such as an echo signal corresponding to any one of the vehicle doors, an echo signal corresponding to a specified vehicle door, or the average of echo signals corresponding to multiple vehicle doors. In an embodiment of the present application, the first signal corresponding to the vehicle door may be determined based on the distance between the emission position of the detection signal and the vehicle door, and / or the distance between the reception position of the echo signal and the vehicle door, as well as the signal reception frequency. For example, the distance resolution may be determined based on the signal reception frequency. For example, if the signal reception frequency is 2 GHz, the distance resolution is 15 cm. The distance between the reception position of the echo signal and the vehicle door is then divided by the distance resolution to obtain the tap number of the echo signal corresponding to the vehicle door, thereby selecting the echo signal corresponding to the tap number as the first signal.

[0106] In some embodiments, step S111 may include: determining the standard deviation of multiple signals in the response signal matrix; and determining the response signal standard deviation based on the standard deviations of the multiple signals. For example, the standard deviation of each signal in the response signal matrix may be determined, and then the response signal standard deviation may be determined based on the standard deviations of all signals in the response signal matrix. In some embodiments, determining the response signal standard deviation based on the standard deviations of the multiple signals may include: averaging the standard deviations of the multiple signals to obtain the response signal standard deviation. Of course, in practical applications, weighted averaging or other processing may also be used to determine the response signal standard deviation, or the standard deviation with the largest or smallest value may be selected from the standard deviations of the multiple signals as the response signal standard deviation.

[0107] In some embodiments, the above step S120 may include the following steps:

[0108] Step S121: determining a second signal according to the response signal matrix;

[0109] Step S122: Determine whether there is a stationary target in the vehicle according to the second signal.

[0110] The second signal may be any one of the signals in the response signal matrix; or, the second signal may be a signal in the response signal matrix that satisfies certain conditions. For example, the second signal is the signal with the largest standard deviation in the response signal matrix. Since the signs of life activities of static targets are not obvious, in order to improve the accuracy of static target detection, the embodiment of the present application may perform data enhancement on the response signal matrix to improve the signal-to-noise ratio of the signal in the response signal matrix. Based on this, in some embodiments, the above step S121 may include the following steps:

[0111] Step S1211: performing enhancement processing on the signals in the response signal matrix to obtain an enhanced response signal matrix;

[0112] Step S1212: Determine a second signal according to the enhanced response signal matrix.

[0113] The second signal may be any one of the signals in the enhanced response signal matrix; or, the second signal may be a signal in the enhanced response signal matrix that satisfies certain conditions. For example, the second signal is the signal with the largest standard deviation in the enhanced response signal matrix. Based on this, in some embodiments, the above step S1212 may include: selecting the second signal from the enhanced response signal matrix. The second signal is the signal with the largest standard deviation in the enhanced response signal matrix. The embodiment of the present application does not limit the specific method of enhancement processing, and in actual application, it can be flexibly set according to needs. For example, enhancement processing includes but is not limited to: bandpass filtering, projection, signal expansion and other processing.

[0114] In some embodiments, the above step S1211 may include: performing bandpass filtering on the signals in the response signal matrix to obtain an enhanced response signal matrix.

[0115] Bandpass filtering can filter out breathing signals from the response signal matrix to improve the efficiency of static target detection. For static targets, such as a sleeping infant, only breathing movements may be present. Furthermore, in certain leakage scenarios, such as a living subject near a vehicle without any movement, the standard deviation variation is weak, making it impossible to determine the presence of a living subject using the aforementioned moving target detection method. Therefore, the response signal matrix needs to be enhanced to filter out effective breathing signals.

[0116] The embodiments of the present application do not limit the specific parameters of the bandpass filtering process, and can be flexibly set according to actual needs in actual applications. For example, since the infant's respiratory rate is approximately 20bpm to 40bpm, that is, the corresponding frequency is approximately 0.3Hz to 0.7Hz, the lower limit frequency of the passband of the bandpass filtering process can be 0.3Hz, and the upper limit frequency of the passband can be 0.7Hz; the upper limit frequency of the stopband can be 1.5Hz, and the lower limit frequency of the stopband can be 0.03Hz; the maximum attenuation of the passband can be 3dB, and the minimum attenuation of the stopband can be 30dB.

[0117] In some embodiments, the above step S1211 may include: performing projection processing on the signals in the response signal matrix to obtain an enhanced response signal matrix.

[0118] Due to different in-car environments, different sleeping positions of infants, or different positions of infants in the car, the reflection points of the detection signal will vary, which in turn causes the position of the corresponding I / Q signal to be random. In poor conditions, the I signal, Q signal, amplitude, and phase will all have limitations, making it impossible to fully characterize the signal changes caused by the breathing of a stationary target (such as an infant), and are also accompanied by problems such as signal superposition. In order to effectively extract the breathing signal of a stationary target (such as an infant), the vector projection method of the I / Q complex plane is used to combine the I and Q signals of the echo signal. The optimal projection signal is further determined by rotating the projection plane, thereby improving the signal-to-noise ratio through software methods.

[0119] Based on this, in some embodiments, the above-mentioned projection processing of the signals in the response signal matrix to obtain an enhanced response signal matrix includes: for each signal in the response signal matrix, obtaining the projection components of the signal on multiple projection planes; selecting the projected signal from the multiple projection components; and generating an enhanced response signal matrix based on the multiple projected signals.

[0120] The projection plane can be a virtual projection plane on the I / Q complex plane, and the projection plane can be considered as a vector on the I / Q complex plane. In order to obtain the projection components of the signal in all directions of the entire I / Q complex plane, the embodiment of the present application can set multiple projection planes to respectively obtain the projection components of the signal on multiple projection planes, that is, multiple projection components. For example, the rotation angle step can be set to 15 degrees, and then the angle of the projection plane can be changed from the range of [0,180] to obtain 12 projection planes. One projection component can be selected from the multiple projection components as the projected signal. The projected signal can be any projection component or a projection component that meets certain requirements. For example, in order to obtain the best projection signal, the projected signal is the projection component with the largest standard deviation among the multiple projection components.

[0121] In some embodiments, the above step S1211 may include: performing signal expansion processing on the response signal matrix to obtain an enhanced response signal matrix.

[0122] Due to the complex environment inside the car, the detection signal is reflected once or multiple times on physical surfaces at different locations, which will cause the multipath effect to continue to accumulate. Although the corresponding physical distance in the car is roughly the distance between the emission point of the detection signal and the car door, it is very difficult to obtain a baby breathing signal with a high signal-to-noise ratio in the corresponding physical distance in the car due to the baby's sleeping posture, the baby seat sitting upright or upside down, and the baby in the footrest area. The embodiment of the present application can use the multipath effect to broaden the Tap numbering interval. For example, the Tap number of the signal in the original response signal matrix is from a to β, and the Tap number can be broadened from the original a to β to a to b, where b is greater than β and β is greater than a.

[0123] In practical applications, when performing enhancement processing, bandpass filtering, projection processing, and signal expansion processing can be combined. For example, bandpass filtering, projection processing, and signal expansion processing are performed in sequence; or, bandpass filtering, signal expansion processing, and projection processing are performed in sequence; or, projection processing, bandpass filtering, and signal expansion processing are performed in sequence; or, projection processing, signal expansion processing, and bandpass filtering are performed in sequence; or, signal expansion processing, projection processing, and bandpass filtering are performed in sequence; or, signal expansion processing, bandpass filtering, and projection processing are used in sequence. Exemplarily, the above-mentioned step S1211 may include: performing bandpass filtering processing on the signals in the response signal matrix to obtain a response signal matrix after bandpass filtering; performing projection processing on the signals in the response signal matrix after bandpass filtering to obtain a projected response signal matrix; performing signal expansion processing on the projected response signal matrix to obtain an expanded response signal matrix. Among them, the bandpass filtering processing can be for the above-mentioned centralized response signal matrix, and the expanded response signal matrix is the enhanced response signal matrix, which can be used for static target detection.

[0124] With signal r″ m For example, the signal r″′ after bandpass filtering m It can be shown as the following formula 11, so the response signal matrix after bandpass filtering can be shown as the following formula 12.

[0125] Formula 11: r″′ m =I″′ m +Q″′ m ×j

[0126] Formula 12: R″′=[r″′1 r″′2 … r″′ m ]

[0127] Among them, r″′ m is the signal r″′ m The real part, Q″′ n is the signal r″′ m The imaginary part of .

[0128] After bandpass filtering, projection processing is performed, and the projection plane is represented by [cosθ, sinθ], then the signal r″′ m The projected component on the projection plane It can be shown as the following formula 13.

[0129] Formula 13:

[0130] After obtaining the projection components of the signal on multiple projection planes, the optimal projection component is selected from the multiple projection components, for example, the projection component with the largest standard deviation is selected as the projected signal. Therefore, the projected response signal matrix can be expressed as follows:

[0131] Formula 14:

[0132] After the projection processing, signal expansion processing is performed to obtain an enhanced response signal matrix.

[0133] The real-time spectrum energy graph confirms that the signal generated by breathing is periodic. Furthermore, a sleeping infant's breathing is periodic and steady. However, when a user stands next to a vehicle, the echo signal obtained is unlikely to be periodic due to physical obstruction by the vehicle's door panels and the addition of body movements. Therefore, periodic detection can be used to determine whether there is breathing (such as a baby) inside the vehicle. It also eliminates situations where the user is close to the vehicle or leaning against the window, thus avoiding false positives in liveness detection.

[0134] In some embodiments, step S122 may include determining a target frequency for the second signal; and determining whether a stationary object is present in the vehicle based on the target frequency. Determining the target frequency for the second signal includes performing spectrum conversion on the second signal to obtain multiple frequencies; and determining the target frequency based on the multiple frequencies. Spectral conversion includes, but is not limited to, Fourier transform, fast Fourier transform, or mel spectrum. For example, performing a fast Fourier transform on the second signal may obtain multiple frequencies, and then determining the target frequency based on the multiple frequencies. The target frequency may be any one of the multiple frequencies, or an average or weighted average of the multiple frequencies. For example, the target frequency may be selected from the multiple frequencies, and the target frequency may be the frequency with the largest modulus among the multiple frequencies. Determining whether a stationary object is present in the vehicle based on the target frequency includes comparing the target frequency with a second threshold to determine whether a stationary object is present in the vehicle. If the target frequency does not exceed the second threshold, a stationary object is present in the vehicle; if the target frequency exceeds the second threshold, no stationary object is present in the vehicle. The second threshold may be a single threshold or multiple thresholds. For example, the second threshold may be two thresholds, one of which may be 0.25 Hz, 0.3 Hz, or 0.32 Hz, and the other may be 0.65 Hz, 0.7 Hz, or 0.73 Hz. Taking the second thresholds of 0.3 Hz and 0.7 Hz as an example, if the target frequency w is greater than or equal to 0.3 Hz and less than or equal to 0.7 Hz, a stationary object exists in the vehicle; if the target frequency w is less than 0.3 Hz or greater than 0.7 Hz, no stationary object exists in the vehicle.

[0135] In some embodiments, the above-mentioned step S122 may include: determining the autocorrelation function of the second signal; and determining whether there is a stationary target in the vehicle based on the autocorrelation function. Determining whether there is a stationary target in the vehicle based on the autocorrelation function includes: determining whether there is a stationary target in the vehicle based on the comparison of the slope of the autocorrelation function and the third threshold. If the second signal is a periodic signal, the lag time delay in the autocorrelation function of the second signal and the corresponding peak value, that is, the slope of the autocorrelation function, should be less than or equal to the third threshold. The third threshold can be flexibly set according to the needs, and the embodiments of the present application do not limit this. If the slope of the autocorrelation function is less than or equal to the third threshold, there is a stationary target in the vehicle; if the slope of the autocorrelation function is greater than the third threshold, there is no stationary target in the vehicle.

[0136] In some embodiments, step S122 may include: determining the primary frequency modulus and secondary frequency modulus of the second signal; and determining whether a stationary target is present in the vehicle based on the primary frequency modulus and secondary frequency modulus. Determining whether a stationary target is present in the vehicle based on the primary frequency modulus and secondary frequency modulus includes: determining whether a stationary target is present in the vehicle based on a comparison of the sum of the ratios of the secondary frequency modulus to the primary frequency modulus with a fourth threshold. For a periodic signal, the modulus corresponding to its primary frequency should be much greater than the modulus of other secondary frequencies. Therefore, the sum of the ratios of all secondary frequency modulus to the primary frequency modulus can be calculated, and this sum of ratios should be less than or equal to the fourth threshold. This fourth threshold can be flexibly set based on needs and is not limited in this embodiment of the present application. If the sum of the ratios of all secondary frequency modulus to the primary frequency modulus is less than or equal to the fourth threshold, then a stationary target is present in the vehicle; if the sum of the ratios of all secondary frequency modulus to the primary frequency modulus is greater than the fourth threshold, then no stationary target is present in the vehicle.

[0137] In practical applications, the frequency, autocorrelation function and modulus value can be used in combination to determine whether there is a static target in the car. For example, the frequency, autocorrelation function and modulus value are used in sequence to determine whether there is a static target in the car; or, the frequency, modulus value and autocorrelation function are used in sequence to determine whether there is a static target in the car; or, the autocorrelation function, frequency and modulus value are used in sequence to determine whether there is a static target in the car; or, the autocorrelation function, modulus value and frequency are used in sequence to determine whether there is a static target in the car; or, the modulus value, autocorrelation function and frequency are used in sequence to determine whether there is a static target in the car; or, the modulus value, frequency and autocorrelation function are used in sequence to determine whether there is a static target in the car. Exemplarily, the above-mentioned step S122 may include: determining the target frequency of the second signal; if the target frequency of the second signal exceeds the second threshold, there is no stationary target in the vehicle; if the target frequency of the second signal does not exceed the second threshold, determining the autocorrelation function of the second signal; if the slope of the autocorrelation function of the second signal is greater than the third threshold, there is no stationary target in the vehicle; if the slope of the autocorrelation function of the second signal is less than or equal to the third threshold, determining the main frequency modulus and the secondary frequency modulus of the second signal; if the sum of the ratios of all secondary frequency modulus values to the main frequency modulus value is less than or equal to the fourth threshold, there is a stationary target in the vehicle; if the sum of the ratios of all secondary frequency modulus values to the main frequency modulus value is greater than the fourth threshold, there is no stationary target in the vehicle.

[0138] In summary, the technical solutions provided by the embodiments of this application achieve targeted liveness detection and improve the accuracy of liveness detection by classifying living objects into moving and stationary targets and providing matching detection methods for each. Furthermore, the combined detection of moving and stationary targets can address the issue of signal leakage outside the vehicle and avoid misjudgment of liveness detection.

[0139] According to a second aspect of the present application, embodiments of the present application further provide a non-transitory computer-readable storage medium having a computer program stored thereon. When executed by a processor, the program implements the steps of the liveness detection method described above. This non-transitory computer-readable storage medium has all the beneficial effects of the liveness detection method described above, and this application will not further elaborate on them.

[0140] According to the third aspect of the present application, an embodiment of the present application further provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the above-mentioned liveness detection method and has all the beneficial effects of the above-mentioned liveness detection method. This application will not go into details here.

[0141] According to a fourth aspect of the present application, an embodiment of the present application further provides an electronic device comprising: a memory and a processor, wherein the memory stores a computer program; the processor is configured to execute the computer program in the memory to implement the steps of the liveness detection method described above. This electronic device has all the beneficial effects of the liveness detection method described above, and this application will not further elaborate on them.

[0142] The computer-readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or any combination thereof, and this application does not specifically limit this. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0143] In some embodiments of the present application, a computer-readable storage medium may be any tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device.

[0144] The computer-readable storage medium may be included in the electronic device or may exist independently without being incorporated into the electronic device. The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device:

[0145] Determine whether there is a living body in the vehicle according to the response signal matrix;

[0146] The response signal matrix includes multipath echo signals of the detection signal.

[0147] Computer program code for performing the operations of some embodiments of the present application can be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer through any type of network (including a local area network (LAN) or a wide area network (WAN)), or can be connected to an external computer (for example, using an Internet service provider to connect via the Internet).

[0148] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of the systems, methods, and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function.

[0149] It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures.

[0150] For example, two blocks shown in succession may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functionality involved. It should also be noted that each block in the block diagrams and / or flow charts, and combinations of blocks in the block diagrams and / or flow charts, may be implemented using a dedicated hardware-based system that performs the specified functions or operations, or may be implemented using a combination of dedicated hardware and computer instructions.

[0151] The units described in some embodiments of the present application may be implemented in software or hardware, and may also be provided in a processor.

[0152] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: Field Programmable Gate Array (FPGA), Application Specific Integrated Circuit (ASIC), Application Specific Standard Parts (ASSP), System on Chip (SOC), Complex Programmable Logic Device (CPLD), and the like.

[0153] According to a fifth aspect of the present application, embodiments of the present application further provide a liveness detection system. The liveness detection system is configured to perform the liveness detection method described above and has all the beneficial effects of the liveness detection method described above, which are not further described herein. The liveness detection system includes a first radar and an electronic device.

[0154] The first radar is used to transmit a detection signal and receive a multipath echo signal of the detection signal.

[0155] The electronic device is used to determine whether a living body exists in the vehicle based on a response signal matrix, wherein the response signal matrix includes multipath echo signals.

[0156] The embodiments of the present application do not limit the frequency at which the first radar transmits the detection signal. This frequency can be flexibly set based on actual needs in practical applications. For example, the sampling frequency of the first radar's transmitting antenna can be 50 Hz, 55 Hz, or 100 Hz. If the sampling frequency of the first radar's transmitting antenna is not 50 Hz, the first radar transmits a detection signal every 20 ms. The embodiments of the present application also do not limit the bandwidth of the detection signal. This frequency can be flexibly set based on actual needs in practical applications. For example, the bandwidth of the detection signal can be 500 MHz, 550 MHz, or 600 MHz.

[0157] The first radar is connected to an electronic device and can transmit the received multipath echo signals to the electronic device. The electronic device can generate a response signal matrix based on the multipath echo signals and then determine whether a living person is present within the vehicle based on the response signal matrix. The specific process for determining whether a living person is present within the vehicle based on the response signal matrix is described in the above embodiment and will not be further elaborated here.

[0158] The embodiments of the present application do not limit the specific number, type, and installation location of the first radar, and can be flexibly set according to needs in actual applications. In some embodiments, the first radar includes one radar, and the type of the first radar is self-receiving and self-transmitting, that is, the first radar includes two antennas, one antenna is used to transmit the detection signal, and the other antenna is used to receive the echo signal of the detection signal; or, the first radar includes two radars, and the type of the first radar is one-receiving and one-transmitting, that is, one radar is used to transmit the detection signal, and the other radar is used to receive the echo signal of the detection signal. Of course, in actual applications, the first radar can also include other quantities or other types, such as the first radar includes three radars, the type is two-receiving and one-transmitting; or, the first radar includes four radars, the type is two-receiving and two-transmitting.

[0159] To ensure a more comprehensive detection range for the first radar, the first radar can be installed at the center of the vehicle. For example, if the first radar comprises one radar, the installation location of the first radar includes any of the following: the center of the roof or the center of the passenger compartment. For example, if the first radar comprises two radars, the installation location of the first radar includes any of the following: the center of the front and rear roofs, the center of the left and right roofs, the center of the front and rear passenger compartments, the center of the left and right passenger compartments. The center of the passenger compartment can be the location of the armrest; the center of the front passenger compartment can be the location of the multimedia device at the front of the passenger compartment; the center of the rear passenger compartment can be the center of the seat at the rear of the passenger compartment; the center of the left passenger compartment can be the center between the left front door and the left rear door; and the center of the right passenger compartment can be the center between the right front and right rear doors. In practical applications, the installation position of the first radar may also have other settings, which should all fall within the scope of protection of this application.

[0160] According to the sixth aspect of this application, Figure 2 As shown, the embodiment of the present application further provides a vehicle 10, which includes the above-mentioned electronic device or the above-mentioned living body detection system. The vehicle has all the beneficial effects of the above-mentioned electronic device and living body detection system, etc., and this application will not repeat them here.

[0161] The vehicle may be a fuel vehicle, a plug-in hybrid vehicle or a new energy vehicle, etc., and this application does not make any specific restrictions on this.

[0162] Below, a specific example is used to introduce and illustrate the liveness detection method and liveness detection system provided in the embodiments of the present application.

[0163] In the embodiment of the present application, liveness detection needs to avoid judging the presence of living things inside the car at a close distance outside the car (for example, when the user closes and locks the car door and leans against the door, or when the user looks inside the car close to the window glass, etc.). This scenario can be called an interference scenario.

[0164] Furthermore, a key aspect of liveness detection is ensuring the presence of an infant in the vehicle. However, the location of an infant can be random. To verify the accuracy of liveness detection methods, standard one-year-old and three-year-old infant models can be used in testing. The test scenarios include: in-vehicle scenarios: a one-year-old infant lying on his back and lying on his stomach in all five seats; a one-year-old infant sitting upright in a rear-seat infant seat; a three-year-old infant sitting upright in all five seats; a three-year-old infant sitting upright in a rear-seat infant seat; and scenarios outside the vehicle: an empty scene; a single or multiple person standing and swaying near the vehicle; and a single or multiple person walking around the vehicle.

[0165] The core of liveness detection is to accurately identify whether there is a living person in the car, and to be able to determine whether there is a living person in the car in interference scenarios. Figure 3 As shown, the coverage of the detection signal emitted by the first radar may exceed the range of in-vehicle liveness detection. To address this technical issue, in-vehicle liveness can be divided into moving and stationary targets. In the case of stationary targets, the breathing signal of a living being is very smooth and exhibits a clear periodicity. Therefore, the periodicity of the breathing signal during sleep can be effectively combined with the periodicity of the sleeping state to determine whether the living being is alive in the vehicle using both the time and frequency domains.

[0166] like Figure 4 As shown, taking the first radar 410 as a one-receive, one-transmit type as an example, the first radar 410 can be installed at the center of the roof. The first radar 410 can be an ultra-wideband radar, thereby avoiding the privacy leakage, poor penetration, and blind spot issues caused by liveness detection through cameras, as well as the high cost of millimeter-wave radar. After the liveness detection conditions are met, the first radar 410 transmits an ultra-wideband signal and receives multipath echo signals of the ultra-wideband signal to generate a response signal matrix. Based on the response signal matrix, the electronic device can perform liveness detection within a first time period and output a detection result. Of course, liveness detection can also continue in time periods after the first time period. For example, the electronic device performs liveness detection and outputs a detection result within 10 seconds, and then performs liveness detection again within 60 seconds and outputs a detection result.

[0167] like Figure 5 As shown, the liveness detection method provided in the embodiment of the present application may include the following steps S501 to S509.

[0168] Step S501: Acquire a response signal matrix, which includes multipath echo signals of a detection signal.

[0169] Step S502: performing outlier filtering on the response signal matrix to obtain an outlier filtered response signal matrix.

[0170] Step S503: performing centralization processing on the response signal matrix after outlier filtering to obtain a centralized response signal matrix.

[0171] Step S504: Determine whether there is a moving target in the vehicle based on the centralized response signal matrix. If there is a moving target, proceed to step S509 below; if not, proceed to step S505 below.

[0172] Step S505: performing band-pass filtering on the centralized response signal matrix to obtain a band-pass filtered response signal matrix.

[0173] Step S506: performing projection processing on the response signal matrix after bandpass filtering to obtain a projected response signal matrix.

[0174] Step S507: performing signal expansion processing on the projected response signal matrix to obtain an expanded response signal matrix.

[0175] Step S508: Determine whether there is a static target in the vehicle based on the expanded response signal matrix. If there is a static target, proceed to step S509 below; if there is no static target, proceed to step S501 above.

[0176] Step S509: Execute the target operation, which includes but is not limited to opening the car window, turning on the air conditioner, and alarming.

[0177] Of course, in actual applications, the types of target operations may vary depending on the type of living object in the vehicle, and this embodiment of the application does not limit this. For example, if there is a moving object in the vehicle, the target operation may include opening the window; if there is a stationary object in the vehicle, the target operation may include turning on the air conditioner and sounding the alarm.

[0178] Compared to related technologies that use on-board cameras, millimeter-wave radars, and carbon dioxide concentration detection to improve the success rate of liveness detection, the embodiments of the present application can achieve full vehicle coverage using only a single UWB radar, reducing costs while avoiding infringements on user privacy. Furthermore, the embodiments of the present application effectively solve the signal superposition problem by adding a virtual multipath method, effectively improving the signal-to-noise ratio of the respiratory signal, and helping to obtain more accurate liveness characteristics. Furthermore, the embodiments of the present application detect moving targets and stationary targets inside the vehicle separately, effectively solving the problem of signal leakage outside the vehicle.

[0179] related Figure 5 For detailed description of each step and its beneficial effects in the embodiment, please refer to the above embodiment and will not be repeated here.

[0180] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0181] In the above embodiments, 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.

[0182] The embodiments, implementation methods and related technical features of the present application can be combined and replaced with each other without conflict.

[0183] The above are merely preferred embodiments of the present application and do not constitute any form of limitation to the present application. Although the descriptions of each embodiment in the embodiments of the present application have different focuses, for parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application are still within the scope of the technical solution of the present application.

Claims

1. A method for detecting a living body, characterized in that: The method comprises: Determine whether there is a living body in the vehicle according to the response signal matrix; The response signal matrix includes multipath echo signals of the detection signal.

2. The living body detection method according to claim 1, characterized in that The detection signal is an ultra-wideband signal.

3. The living body detection method according to claim 1, characterized in that: The signals in the response signal matrix are channel impulse response signals.

4. The method for detecting living body according to claim 1, wherein: The determining whether there is a living body in the vehicle according to the response signal matrix includes: Determine whether there is a moving target in the vehicle according to the response signal matrix; wherein the living body includes the moving target.

5. The method for liveness detection according to claim 4, wherein: Determining whether there is a moving target in the vehicle according to the response signal matrix includes: determining a response signal standard deviation based on a response signal matrix; Determine whether there is a moving target in the vehicle according to the standard deviation of the response signal.

6. The living body detection method according to claim 5, characterized in that: Determining the response signal standard deviation according to the response signal matrix includes: selecting a first signal from the response signal matrix; A standard deviation of a response signal is determined based on the first signal.

7. The living body detection method according to claim 6, characterized in that: The first signal is an echo signal corresponding to the vehicle door.

8. The living body detection method according to claim 5, characterized in that: Determining the response signal standard deviation according to the response signal matrix includes: determining the standard deviation of multiple signals in a response signal matrix; A response signal standard deviation is determined based on the standard deviations of the plurality of signals.

9. The living body detection method according to claim 8, characterized in that: Determining the standard deviation of the response signal according to the standard deviations of the plurality of signals includes: The standard deviations of the plurality of signals are averaged to obtain a response signal standard deviation.

10. The living body detection method according to claim 5, characterized in that: The determining whether there is a moving target in the vehicle according to the standard deviation of the response signal includes: It is determined whether there is a moving target in the vehicle based on a comparison between the response signal standard deviation and a first threshold.

11. The living body detection method according to claim 1, characterized in that: The determining whether there is a living body in the vehicle according to the response signal matrix includes: Determine whether there is a static target in the vehicle according to the response signal matrix; wherein the living body includes the static target.

12. The living body detection method according to claim 11, characterized in that: Determining whether there is a static target in the vehicle according to the response signal matrix includes: determining a second signal according to the response signal matrix; Determine whether there is a stationary target in the vehicle according to the second signal.

13. The living body detection method according to claim 12, characterized in that: The determining the second signal according to the response signal matrix includes: performing enhancement processing on the signals in the response signal matrix to obtain the enhanced response signal matrix; A second signal is determined according to the enhanced response signal matrix.

14. The living body detection method according to claim 13, characterized in that: The step of performing enhancement processing on the signals in the response signal matrix to obtain the enhanced response signal matrix includes: Bandpass filtering is performed on the signals in the response signal matrix to obtain the enhanced response signal matrix.

15. The living body detection method according to claim 13, characterized in that: The step of performing enhancement processing on the signals in the response signal matrix to obtain the enhanced response signal matrix includes: Projection processing is performed on the signals in the response signal matrix to obtain the enhanced response signal matrix.

16. The living body detection method according to claim 15, characterized in that: The projecting process is performed on the signal in the response signal matrix to obtain the enhanced response signal matrix, including: For each signal in the response signal matrix, obtaining projection components of the signal on multiple projection planes; selecting the projected signal from a plurality of the projected components; An enhanced response signal matrix is generated according to the projected multiple signals.

17. The living body detection method according to claim 13, characterized in that: The step of performing enhancement processing on the signals in the response signal matrix to obtain the enhanced response signal matrix includes: Signal expansion processing is performed on the response signal matrix to obtain the enhanced response signal matrix.

18. The living body detection method according to claim 13, characterized in that: The determining the second signal according to the enhanced response signal matrix includes: A second signal is selected from the enhanced response signal matrix; wherein the second signal is a signal with the largest standard deviation in the enhanced response signal matrix.

19. The living body detection method according to claim 12, wherein: The determining whether there is a stationary target in the vehicle according to the second signal includes: determining a target frequency of the second signal; It is determined whether there is a static target in the vehicle according to the target frequency.

20. The living body detection method according to claim 19, characterized in that: The determining a target frequency of the second signal includes: performing spectrum conversion processing on the second signal to obtain multiple frequencies; A target frequency is determined based on the plurality of frequencies.

21. The living body detection method according to claim 20, characterized in that: The determining of the target frequency according to the multiple frequencies includes: A target frequency is selected from the multiple frequencies; wherein the target frequency is the frequency with the largest modulus value among the multiple frequencies.

22. The living body detection method according to claim 19, characterized in that: Determining whether there is a stationary target in the vehicle according to the target frequency includes: According to the comparison between the target frequency and the second threshold, it is determined whether there is a stationary target in the vehicle.

23. The living body detection method according to claim 12, characterized in that: The determining whether there is a stationary target in the vehicle according to the second signal includes: determining an autocorrelation function of the second signal; It is determined whether there is a static target in the vehicle according to the autocorrelation function.

24. The living body detection method according to claim 23, characterized in that: Determining whether there is a static target in the vehicle according to the autocorrelation function includes: It is determined whether there is a stationary target in the vehicle based on a comparison between the slope of the autocorrelation function and a third threshold.

25. The living body detection method according to claim 12, characterized in that: The determining whether there is a stationary target in the vehicle according to the second signal includes: determining a primary frequency modulus value and a secondary frequency modulus value of the second signal; Determine whether there is a static target in the vehicle according to the main frequency modulus value and the secondary frequency modulus value.

26. The living body detection method according to claim 25, characterized in that: The determining whether there is a static target in the vehicle according to the main frequency modulus value and the secondary frequency modulus value includes: Whether there is a stationary target in the vehicle is determined based on a comparison between the sum of the ratios of the secondary frequency modulus to the primary frequency modulus and a fourth threshold.

27. The living body detection method according to claim 1, characterized in that: The method further comprises: Performing data preprocessing on the response signal matrix to obtain the preprocessed response signal matrix; wherein the preprocessed response signal matrix is used to determine whether there is a living body in the vehicle.

28. The living body detection method according to claim 27, characterized in that: The performing data preprocessing on the response signal matrix to obtain the preprocessed response signal matrix includes: Perform outlier filtering on the response signal matrix to obtain the preprocessed response signal matrix.

29. The living body detection method according to claim 28, characterized in that: The performing outlier filtering on the response signal matrix to obtain the preprocessed response signal matrix includes: For each signal in the response signal matrix, determining the preprocessed signal based on a comparison between the signal and the median absolute deviation of the window in which the signal is located; The pre-processed response signal matrix includes a plurality of pre-processed signals.

30. The living body detection method according to claim 27, characterized in that: The performing data preprocessing on the response signal matrix to obtain the preprocessed response signal matrix includes: Centralizing the response signal matrix to obtain the pre-processed response signal matrix.

31. The living body detection method according to claim 30, characterized in that: The centralizing the response signal matrix to obtain the pre-processed response signal matrix includes: determining the signal mean in the response signal matrix; Each signal in the response signal matrix is subtracted from the signal mean to obtain the preprocessed response signal matrix.

32. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the living body detection method according to any one of claims 1 to 31 is implemented.

33. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the living body detection method according to any one of claims 1 to 31 is implemented.

34. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the liveness detection method according to any one of claims 1 to 31.

35. A living body detection system, characterized in that: The living body detection system includes a first radar and an electronic device; wherein, The first radar is used to: transmit a detection signal and receive a multipath echo signal of the detection signal; The electronic device is used to determine whether a living body exists in a vehicle according to a response signal matrix; wherein the response signal matrix includes the multipath echo signal.

36. The living body detection system according to claim 35, characterized in that: The type of the first radar is self-transmitting and self-receiving.

37. The living body detection system according to claim 36, characterized in that: The installation position of the first radar includes any one of the following: the center of the roof, the center of the passenger compartment.

38. The living body detection system according to claim 35, characterized in that: The first radar includes two radars, and the type of the first radar is one receiving and one transmitting.

39. The living body detection system according to claim 38, characterized in that: The installation position of the first radar includes any one of the following: the center position of the front side of the roof and the center position of the rear side of the roof, the center position of the left side of the roof and the center position of the right side of the roof, the center position of the front side of the passenger compartment and the center position of the rear side of the passenger compartment, the center position of the left side of the passenger compartment and the center position of the right side of the passenger compartment.

40. A vehicle, characterized in that: The device comprises the electronic device as claimed in claim 34, or the living body detection system as claimed in any one of claims 35 to 39.

Images