Living body detection method and related equipment

Abstract

The embodiment of the invention discloses a living body detection method and related equipment, which are used for carrying out living body detection under the condition of improving the accuracy of living body detection. The method provided by the embodiment of the invention comprises the following steps: if a trigger signal of living body detection is detected, controlling a Bluetooth transmitting device to periodically transmit a first signal, and obtaining a signal strength indication of the first signal received by a Bluetooth receiving device, determining whether a living body exists in the target environment of the target environment factor based on the distribution condition of the signal intensity indication of the first signal, if so, controlling an emitter of the radar device to emit a second signal, and obtaining echo data of the second signal received by a receiver of the radar device, and inputting the echo data of the second signal into a pre-trained living body detection model, and performing living body detection of a target environment of the target environment factor on the echo data of the second signal by the living body detection model to obtain a living body detection result corresponding to the echo data of the second signal.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of liveness detection, and more specifically, to a liveness detection method, a liveness detection apparatus, a liveness detection device, and a computer-readable storage medium. Background Art

[0002] As cars become more intelligent and automated, in-car occupant detection becomes increasingly important. Traditional detection technologies, such as infrared sensors and cameras, have problems such as low detection accuracy, being greatly affected by the environment, and being controversial in terms of privacy protection.

[0003] The existing liveness detection method is to equip the car with Bluetooth sensor nodes (positioning anchor points for non-sensing positioning). The wavelength of the 2.4GHz frequency band of Bluetooth is about 12.5 cm. Electromagnetic waves of this wavelength are relatively stable when propagating in the air, but when in contact with the human body, since the human body is mainly composed of water and salt, the propagation of electromagnetic waves will be significantly affected. For human tissue, 2.4GHz electromagnetic waves have a certain penetration ability, but scattering and absorption will occur, resulting in the weakening of the Bluetooth signal strength indicator (RSSI, Received Signal Strength Indicator), that is, the phenomenon of "signal attenuation", which reduces the ability to accurately detect liveness through signal changes, thereby reducing the accuracy of liveness detection.

[0004] Therefore, the liveness detection accuracy of existing liveness detection methods is low. Summary of the invention

[0005] The embodiments of the present application provide a liveness detection method, a liveness detection apparatus, a liveness detection device, and a computer-readable storage medium, which can perform liveness detection while improving the accuracy of liveness detection.

[0006] In a first aspect, an embodiment of the present application provides a living body detection method, comprising:

[0007] If a trigger signal for liveness detection is detected, the Bluetooth transmitting device is controlled to periodically transmit a first signal, and a signal strength indication of the first signal received by the Bluetooth receiving device is obtained;

[0008] Determine whether there is a living body in a target environment of a target environmental factor based on a distribution of the signal strength indication of the first signal, wherein the target environmental factor includes at least one of temperature, humidity and light;

[0009] If so, controlling the transmitter of the radar device to transmit a second signal, and acquiring echo data of the second signal received by the receiver of the radar device;

[0010] The echo data of the second signal is input into a pre-trained liveness detection model, and the liveness detection model performs liveness detection of the target environment of the target environment factor on the echo data of the second signal to obtain a liveness detection result corresponding to the echo data of the second signal.

[0011] In a second aspect, an embodiment of the present application provides a living body detection device, including:

[0012] an acquisition unit, configured to control the Bluetooth transmitting device to periodically transmit a first signal if a trigger signal for liveness detection is detected, and to acquire a signal strength indication of the first signal received by the Bluetooth receiving device;

[0013] A determination unit, configured to determine whether there is a living body in a target environment of a target environmental factor based on a distribution of a signal strength indication of the first signal, wherein the target environmental factor includes at least one of temperature, humidity and light;

[0014] The acquisition unit is further configured to, if present, control the transmitter of the radar device to transmit a second signal, and acquire echo data of the second signal received by the receiver of the radar device;

[0015] A model detection unit is used to input the echo data of the second signal into a pre-trained liveness detection model, and the liveness detection model performs liveness detection of the target environment of the target environmental factors on the echo data of the second signal to obtain a liveness detection result corresponding to the echo data of the second signal.

[0016] In a third aspect, an embodiment of the present application provides a living body detection device, including:

[0017] CPU, memory, input and output interfaces, wired or wireless network interfaces, and power supply;

[0018] The memory is a short-term storage memory or a persistent storage memory;

[0019] The central processing unit is configured to communicate with the memory and execute instruction operations in the memory to perform the aforementioned living body detection method.

[0020] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which includes instructions. When the instructions are executed on a computer, the computer executes the aforementioned liveness detection method.

[0021] In a fifth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the aforementioned liveness detection method.

[0022] It can be seen from the above technical scheme that the embodiment of the present application has the following advantages: it can determine whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal. If there is, the transmitter of the radar device is controlled to transmit the second signal, and the echo data of the second signal received by the receiver of the radar device is obtained, and the echo data of the second signal received by the live body detection model is used to detect the live body of the target environment of the target environmental factor. The Bluetooth RSSI data (the signal strength indication of the first signal) can provide real-time signal strength change information, and the UWB radar data (the echo data of the second signal) can provide high-precision location information and motion detection. By combining the Bluetooth RSSI data and the UWB radar data, the accuracy of live body detection is high. Further, by monitoring and considering the target environmental factors (such as temperature, humidity and light, etc.), the live body detection strategy can be adjusted according to the actual situation, the influence of environmental factors on the accuracy of live body detection can be reduced, the dependence on environmental conditions can be reduced, the robustness and stability of the system can be improved, and the reliability of live body detection can be high. Furthermore, it is possible to first determine whether there is a living organism in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal. If so, the transmitter of the radar device is controlled to transmit a second signal, and the echo data of the second signal received by the receiver of the radar device is obtained, and the echo data of the second signal is input into a pre-trained liveness detection model, and the liveness detection is performed by the liveness detection model. Therefore, the effect of both high liveness detection accuracy and low system loss can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic diagram of the architecture of a liveness detection system disclosed in an embodiment of the present application;

[0024] Figure 2 A schematic diagram of a flow chart of a liveness detection method disclosed in an embodiment of the present application;

[0025] Figure 3 A schematic diagram of a flow chart of another liveness detection method disclosed in an embodiment of the present application;

[0026] Figure 4 A schematic diagram of the content of a first signal transmitted by a Bluetooth transmitting device disclosed in an embodiment of the present application;

[0027] Figure 5 A schematic diagram of a UWB radar signal processing logic disclosed in an embodiment of the present application;

[0028] Figure 6 A schematic diagram of a living body determination waveform data disclosed in an embodiment of the present application;

[0029] Figure 7 A schematic diagram of the hardware structure of a living body detection system disclosed in an embodiment of the present application;

[0030] Figure 8 A schematic diagram of the software structure of a liveness detection system disclosed in an embodiment of the present application;

[0031] Fig. 9 A schematic diagram of a moving average filtering of Bluetooth RSSI data disclosed in an embodiment of the present application;

[0032] Fig.10 A schematic diagram of a CWT of a respiratory signal disclosed in an embodiment of the present application;

[0033] Fig.11 A schematic diagram of HHT of a respiratory signal disclosed in an embodiment of the present application;

[0034] Fig.12 A schematic diagram of Kalman filter data fusion disclosed in an embodiment of the present application;

[0035] Fig.13 A schematic diagram of a Kalman filter estimation result disclosed in an embodiment of the present application;

[0036] Fig.14 A schematic diagram of the structure of a living body detection device disclosed in an embodiment of the present application;

[0037] Fig.15 A schematic diagram of the structure of another living body detection device disclosed in an embodiment of the present application;

[0038] Fig.16 A schematic diagram of the structure of a living body detection device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0039] The embodiments of the present application provide a liveness detection method, a liveness detection apparatus, a liveness detection device, and a computer-readable storage medium, which can perform liveness detection while improving the accuracy of liveness detection.

[0040] See also Figure 1 , the architecture of the liveness detection system in the embodiment of the present application includes:

[0041] Liveness detection device 101 and target environment control system 102. When performing liveness detection, the liveness detection device 101 can be connected to the target environment control system 102. If the liveness detection device 101 detects a trigger signal for liveness detection, the Bluetooth transmitting device can be controlled to periodically transmit a first signal, and a signal strength indication of the first signal received by the Bluetooth receiving device can be obtained. Based on the distribution of the signal strength indication of the first signal, it can be determined whether there is a live body in the target environment of the target environmental factor. If there is, the transmitter of the radar device can be controlled to transmit a second signal, and the echo data of the second signal received by the receiver of the radar device can be obtained. The liveness detection model performs liveness detection of the target environment of the target environmental factor on the echo data of the second signal, and obtains the liveness detection result corresponding to the echo data of the second signal, and sends the liveness detection result to the target environment control system 102.

[0042] based on Figure 1 See the liveness detection system shown in Figure 2 , Figure 2 This is a flow chart of a liveness detection method disclosed in an embodiment of the present application, the method comprising:

[0043] 201. If a trigger signal for liveness detection is detected, control a Bluetooth transmitting device to periodically transmit a first signal, and obtain a signal strength indication of the first signal received by a Bluetooth receiving device.

[0044] In this embodiment, when liveness detection is performed, if a trigger signal for liveness detection is detected, the Bluetooth transmitting device may be controlled to periodically transmit a first signal, and a signal strength indication of the first signal received by the Bluetooth receiving device may be obtained.

[0045] It is understandable that the scenarios for liveness detection may include but are not limited to the interior of a car (in-car), a nursing home, and other reasonable scenarios where liveness detection can be performed.

[0046] Specifically, for example, assuming that the application scenario of the embodiment of the present application is a liveness detection scenario in a car, the Bluetooth transmitter can be set at the front position or hand gear position or other specific positions in the car that can provide a larger signal transmission range, and the Bluetooth receiving device can be set at the ceiling light position in the car. The Bluetooth receiving device can periodically scan the broadcast emitted by the Bluetooth transmitter every 100ms.

[0047] 202. Determine whether there is a living body in a target environment of a target environmental factor based on a distribution of the signal strength indication of the first signal, where the target environmental factor includes at least one of temperature, humidity, and light.

[0048] After obtaining the signal strength indication of the first signal received by the Bluetooth receiving device, it can be determined whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal. The target environmental factor includes at least one of temperature, humidity and light.

[0049] It should be understood that Bluetooth signals are electromagnetic waves, which are affected by environmental factors such as temperature, humidity and light. That is, the distribution of the signal strength indication of the first signal corresponding to different environmental factors such as temperature, humidity and light will be different. Therefore, taking the target environmental factors (such as temperature, humidity and light) into consideration can reduce the impact of environmental factors on the accuracy of liveness detection and reduce dependence on environmental conditions, thereby improving the accuracy of liveness detection.

[0050] It is worth mentioning that determining whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal can improve the accuracy of live body detection in the target environment of different target environmental factors.

[0051] 203. If yes, control the transmitter of the radar device to transmit the second signal, and obtain echo data of the second signal received by the receiver of the radar device.

[0052] After determining whether there is a living body in the target environment of the target environment factor based on the distribution of the signal strength indication of the first signal, if so, the transmitter of the radar device can be controlled to transmit a second signal, and the echo data of the second signal received by the receiver of the radar device can be obtained.

[0053] Specifically, the radar device can also be set at other reasonable locations such as the ceiling light position in the car, and the Bluetooth receiving device and the radar device can be integrated with the ceiling light controller, or the Bluetooth receiving device and the radar device can not be integrated with the ceiling light controller, which is not limited here.

[0054] 204. Input the echo data of the second signal into a pre-trained liveness detection model, and use the liveness detection model to perform liveness detection of a target environment of a target environment factor on the echo data of the second signal to obtain a liveness detection result corresponding to the echo data of the second signal.

[0055] After obtaining the echo data of the second signal received by the receiver of the radar device, the echo data of the second signal can be input into a pre-trained liveness detection model, and the liveness detection model performs liveness detection of the target environment factors on the echo data of the second signal to obtain a liveness detection result corresponding to the echo data of the second signal.

[0056] It is worth mentioning that determining whether the echo data of the second signal has a preset liveness feature of the target environment of the target environment factor can improve the accuracy of liveness detection in target environments with different target environment factors.

[0057] In the embodiment of the present application, it is possible to determine whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal. If there is, the transmitter of the radar device is controlled to transmit a second signal, and the echo data of the second signal received by the receiver of the radar device is obtained, and the liveness detection model performs liveness detection of the target environment of the target environmental factor on the echo data of the second signal. The Bluetooth RSSI data (the signal strength indication of the first signal) can provide real-time signal strength change information, and the UWB radar data (the echo data of the second signal) can provide high-precision location information and motion detection. By combining the Bluetooth RSSI data and the UWB radar data, the accuracy of liveness detection is high. Further, by monitoring and considering the target environmental factors (such as temperature, humidity, and light, etc.), the liveness detection strategy can be adjusted according to the actual situation, the influence of environmental factors on the accuracy of liveness detection can be reduced, the dependence on environmental conditions can be reduced, the robustness and stability of the system can be improved, and the reliability of liveness detection can be high. Furthermore, it is possible to first determine whether there is a living organism in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal. If so, the transmitter of the radar device is controlled to transmit a second signal, and the echo data of the second signal received by the receiver of the radar device is obtained, and the echo data of the second signal is input into a pre-trained liveness detection model, and the liveness detection is performed by the liveness detection model. Therefore, the effect of both high liveness detection accuracy and low system loss can be achieved.

[0058] In the embodiment of the present application, there are multiple methods for determining whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal, and for performing liveness detection of the target environment of the target environmental factor on the echo data of the second signal by the liveness detection model to obtain the liveness detection result corresponding to the echo data of the second signal, based on Figure 2 The liveness detection method shown in the figure, one of which is described below.

[0059] In this embodiment, when liveness detection is performed, if a trigger signal for liveness detection is detected, the Bluetooth transmitting device may be controlled to periodically transmit a first signal, and a signal strength indication of the first signal received by the Bluetooth receiving device may be obtained.

[0060] It is understandable that the scenarios for liveness detection may include but are not limited to the interior of a car (in-car), a nursing home, and other reasonable scenarios where liveness detection can be performed.

[0061] Specifically, for example, assuming that the application scenario of the embodiment of the present application is a liveness detection scenario in a car, the Bluetooth transmitter can be set at the front position or hand gear position or other specific positions in the car that can provide a larger signal transmission range, and the Bluetooth receiving device can be set at the ceiling light position in the car. The Bluetooth receiving device can periodically scan the broadcast emitted by the Bluetooth transmitter every 100ms.

[0062] For more details, please refer to Figure 3 , Figure 3 A schematic diagram of another liveness detection method disclosed in an embodiment of the present application, comprising Figure 3 It can be seen that when the vehicle is started and locked (a trigger signal for liveness detection is detected), the Bluetooth RSSI reception can be started after the Bluetooth sensor in the vehicle starts sending broadcasts, that is, the Bluetooth signal strength indicator (RSSI, Received Signal Strength Indicator) is started to be monitored.

[0063] It is understandable that RSSI is a wireless communication technology indicator used to measure the strength of Bluetooth signals between devices. Generally, the lower the RSSI value, the better the signal quality and the closer the distance.

[0064] After obtaining the signal strength indication of the first signal received by the Bluetooth receiving device, it can be determined whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal. The target environmental factor includes at least one of temperature, humidity and light.

[0065] It is worth mentioning that Bluetooth signals are electromagnetic waves, which are affected by environmental factors such as temperature, humidity and light. That is, the distribution of the signal strength indication of the first signal corresponding to different environmental factors such as temperature, humidity and light will be different. Therefore, taking the target environmental factors (such as temperature, humidity and light) into consideration can reduce the impact of environmental factors on the accuracy of liveness detection and reduce dependence on environmental conditions, thereby improving the accuracy of liveness detection.

[0066] Among them, the method for obtaining the signal strength indication of the first signal received by the Bluetooth receiving device may be, obtaining the signal strength indication of the initial signal, and determining the first signal sent by the Bluetooth sending module in the initial signal, the initial signal including at least one signal transmitted by the Bluetooth sending device; and wherein the method for determining whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal may be, first performing a first filtering process on the first signal to obtain the first signal after the first filtering process, and then determining whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal after the filtering process.

[0067] For details, please continue to refer to Figure 3 ,Depend on Figure 3 It can be seen that after starting the Bluetooth RSSI reception, Bluetooth RSSI data filtering and data processing can be performed. Specifically, the first signal transmitted by the Bluetooth transmitting device (the broadcast content transmitted by the Bluetooth sensor node) can be found in Figure 4 , Figure 4 A schematic diagram of the content of a first signal transmitted by a Bluetooth transmitting device disclosed in an embodiment of the present application, comprising Figure 4 It can be seen that the Bluetooth sensor number in the broadcast and the last two bytes of the VIN code can be used to determine whether it is broadcasting from the vehicle. If it is broadcasting from the vehicle, the RSSI value in the corresponding Bluetooth broadcast frame can be parsed. RSSI indicates the strength of the received signal and is usually expressed as a negative value. The smaller the value, the weaker the signal.

[0068] More specifically, the received RSSI data can be filtered to remove noise. The filtering algorithm uses a sliding average filtering algorithm. For details, please refer to the following formula 1:

[0069]

[0070] Where y[n] is the filtered output signal, x[n] is the input signal, and N is the size of the sliding window, which is set to 5 by default.

[0071] Among them, the method for determining whether there is a living organism in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal may be to determine whether the fluctuation change range of the signal strength indication of the first signal is within the preset living organism fluctuation change range corresponding to the target environment of the target environmental factor, and determine whether there is a living organism in the target environment of the target environmental factor based on the judgment result; and / or determine whether the intensity value range of the signal strength indication of the first signal is within the preset living organism intensity value range corresponding to the target environment of the target environmental factor, and determine whether there is a living organism in the target environment of the target environmental factor based on the judgment result.

[0072] Specifically, for example, assuming that the signal strength indication of the first signal is the following RSSI data (unit: dBm): [-70, -72, -68, -71, -69, -73, -70, -72, -69, -71], determine the average RSSI value of the signal strength indication of the first signal, the average RSSI value = [-70 + (-72) + (-68) + (-71) + (-69) + (-73) + (-70) + (-72) + (-69) + (-71)] / 10 = -70. 5, and the preset range of the fluctuation of living body corresponding to the target environment of the target environment factor is [-3, +3], and / or the preset range of the intensity of living body corresponding to the target environment of the target environment factor is [-73.5, -67.5], wherein the preset range of the intensity of living body corresponding to the target environment of the target environment factor is the sum of the preset range of the fluctuation of living body corresponding to the target environment of the target environment factor and the average RSSI value of the signal strength indication of the first signal, then the following two methods can be used to determine whether there is a living body in the target environment:

[0073] (1) Determine the fluctuation range of the signal strength indication of the first signal, the upper threshold of the fluctuation range = -70.5-(-73) = +2.5, the lower threshold = -70.5-(-68) = -2.5, that is, the fluctuation range is [-2.5, +2.5]. Since the fluctuation range [-2.5, +2.5] of the signal strength indication of the first signal is within the preset living body fluctuation range [-3, +3] corresponding to the target environment of the target environmental factor, it is determined that there is a living body in the target environment.

[0074] (2) Determine the strength value range of the signal strength indication of the first signal, the upper threshold of the strength value range = average RSSI value + 3 = -70.5 + 3 = -67.5, the lower threshold = average RSS value - 3 = -70.5 - 3 = -73.5, that is, the strength value range is [-67.5, -73.5]. Since the strength value range [-73, -68] of the signal strength indication of the first signal is within the preset living body strength value range [-73.5, -67.5] corresponding to the target environment of the target environmental factor, it is determined that there is a living body in the target environment.

[0075] It should be understood that the preset liveness fluctuation change range and the preset liveness intensity value range corresponding to the target environment of different target environmental factors will be different. Therefore, determining whether the fluctuation change range indicated by the signal strength of the first signal is within the preset liveness fluctuation change range and / or the preset liveness intensity value range corresponding to the target environment of the target environmental factor can improve the accuracy of liveness detection in the target environment of different target environmental factors.

[0076] After determining whether there is a living body in the target environment of the target environment factor based on the distribution of the signal strength indication of the first signal, if so, the transmitter of the radar device can be controlled to transmit a second signal, and the echo data of the second signal received by the receiver of the radar device can be obtained.

[0077] Specifically, the radar device can also be set at other reasonable locations such as the ceiling light position in the car, and the Bluetooth receiving device and the radar device can be integrated with the ceiling light controller, or the Bluetooth receiving device and the radar device can not be integrated with the ceiling light controller, which is not limited here.

[0078] For more details, please continue to refer to Figure 3 ,Depend on Figure 3 It can be seen that after Bluetooth RSSI data filtering and data processing, Bluetooth liveness detection can be used to determine whether there is a live body. If not, Bluetooth RSSI data filtering and data processing will continue. If yes, the UWB radar will be started to perform liveness detection based on the radar echo signal (echo data of the second signal) later. That is, after determining the presence of a live body through Bluetooth, the UWB pulse Doppler radar (radar device) can be turned on to detect the in-vehicle environment (target environment) to use the Doppler effect of the radar echo signal for liveness detection. For the specific processing logic of the UWB radar signal, please refer to Figure 5 , Figure 5 A schematic diagram of a UWB radar signal processing logic disclosed in an embodiment of the present application, comprising Figure 5 It can be seen that the reflected radar signal is expressed in the form of CIR (Channel Impulse Response) channel impulse response. CIR is based on the measurement of the receiver voltage and is an equivalent complex number representation of the sampling channel. The real part of CIR represents the envelope amplitude of the received signal, and the imaginary part of CIR represents the received phase signal. The obtained CIR estimate is composed of correlation values ​​obtained at different time points. Each time point in the CIR estimate is called a CIR tap. Each CIR tap contains 2 real bytes (signed int16) and 2 imaginary bytes (signed int16) of orthogonal phase values. By accumulating CIR in the fast-slow time dimension, liveness judgment waveform data can be formed. For specific liveness judgment waveform data, please refer to Figure 6 , Figure 6 is a schematic diagram of a living body judgment waveform data disclosed in an embodiment of the present application, Figure 6 It can be seen that Figure 6 Including X-fasttime (fast time axis), Ys low time (slow time axis) and Z-CIR AMP (channel impulse response amplitude), this three-dimensional information can be used to analyze real-time radar data for accurate living body detection.

[0079] After obtaining the echo data of the second signal received by the receiver of the radar device, the echo data of the second signal can be input into a pre-trained liveness detection model, and the liveness detection model performs liveness detection of the target environment factors on the echo data of the second signal to obtain a liveness detection result corresponding to the echo data of the second signal.

[0080] It is worth mentioning that determining whether the echo data of the second signal has a preset liveness feature of the target environment of the target environment factor can improve the accuracy of liveness detection in target environments with different target environment factors.

[0081] Among them, the method of performing liveness detection on the echo data of the second signal of the target environment of the target environmental factor to obtain the liveness detection result corresponding to the echo data of the second signal may be to determine whether the echo data of the second signal has a preset liveness feature of the target environment of the target environmental factor; if so, determine that the liveness detection result corresponding to the echo data of the second signal is a liveness in the target environment with the target environmental factor; if not, determine that the liveness detection result corresponding to the echo data of the second signal is a liveness in the target environment without the target environmental factor.

[0082] It is worth mentioning that determining whether the echo data of the second signal has a preset living feature of the target environment of the target environment factor can improve the accuracy of determining the living body detection result corresponding to the echo data of the second signal.

[0083] Among them, the echo data of the second signal is input into a pre-trained liveness detection model, and the liveness detection model performs liveness detection of the target environment of the target environment factors on the echo data of the second signal, and the method for obtaining the liveness detection result corresponding to the echo data of the second signal can be: performing a second filtering process on the echo data of the second signal to obtain the echo data of the second signal after the second filtering process; if it is determined that there is a living organism in the target environment based on the echo data of the second signal after the second filtering process, the echo data of the second signal after the second filtering process is input into the liveness detection model, and the liveness detection model performs liveness detection of the target environment of the target environment factors on the echo data of the second signal after the second filtering process to obtain the liveness detection result corresponding to the echo data of the second signal after the second filtering process.

[0084] It is worth mentioning that whether there is a living body in the target environment is first determined based on the echo data of the second signal after the second filtering process. If there is, the echo data of the second signal after the second filtering process is input into the living body detection model to perform living body detection through the living body detection model. Therefore, the effect of both high living body detection accuracy and low system loss can be achieved.

[0085] Among them, the method of performing a second filtering process on the echo data of the second signal to obtain the echo data of the second signal after the second filtering process may be to first extract multi-dimensional features of the echo data of the second signal to obtain the multi-dimensional features of the echo data of the second signal, and then fuse the multi-dimensional features of the echo data of the second signal to obtain fused features, wherein the fused features are the echo data of the second signal after the second filtering process.

[0086] Among them, the method for extracting multi-dimensional features of the echo data of the second signal to obtain the multi-dimensional features of the echo data of the second signal may be: performing time-frequency analysis on the echo data of the second signal according to a continuous wavelet transform algorithm to obtain time-frequency dimensional features of the echo data of the second signal, and the time-frequency dimensional features of the echo data of the second signal are the multi-dimensional features of the echo data of the second signal; and / or performing amplitude and phase analysis on the echo data of the second signal according to a Hilbert transform algorithm to obtain amplitude and phase dimensional features of the echo data of the second signal, and the amplitude and phase dimensional features of the echo data of the second signal are the multi-dimensional features of the echo data of the second signal.

[0087] For details, please continue to refer to Figure 3 ,Depend on Figure 3 It can be seen that after starting the UWB radar and after the radar CIR transmits data, UWB radar filtering and data processing can be performed to determine whether there is a live body in the UWB liveness detection. If not, UWB radar filtering and data processing will continue. If so, the live body information will be reported to the car main control.

[0088] More specifically, the signal processing, data fusion and machine learning classification stages can be performed in sequence. For details, please refer to the following description:

[0089] (1) For the signal processing stage, continuous wavelet transform (CWT) can be used for time-frequency analysis to extract characteristic signals such as frequency, amplitude and phase. For the specific mathematical formula, please refer to the following formula 2:

[0090]

[0091] Among them, X(a,b) is the coefficient of wavelet transform; x(t) is the input signal; ψ is the wavelet function, which is usually selected as the mother wavelet function, using 'cmor' (complex Morlet wavelet); a is the scale parameter that controls the expansion and contraction of the wavelet, using 1 to 128; b is the translation parameter that controls the position of the wavelet, which is determined by the time axis t; ψ* represents the complex conjugate of the wavelet function, specifically the built-in complex conjugate of the 'cmor' wavelet function.

[0092] The instantaneous amplitude and instantaneous phase of the signal can also be analyzed using the Hilbert Transform. For the specific mathematical formula, please refer to the following formula 3:

[0093]

[0094] Where H{x(t)} is the result of Hilbert transform; PV represents the Cauchy principal value of the integral;

[0095] Calculation of instantaneous amplitude and instantaneous phase:

[0096]

[0097] Calculation of instantaneous frequency:

[0098]

[0099] (2) In the data fusion stage, a Kalman filter can be used to fuse multi-sensor data to generate a fused feature vector. For specific mathematical formulas, please refer to the following formulas 6 and 7:

[0100] Prediction Steps:

[0101]

[0102] P k|k-1 =FP k-1|k-1 F T +Q Formula 6

[0103] Update steps:

[0104] K k =P k|k-1 H T (HP k|k-1 H T +R) -1

[0105]

[0106] P k|k =(IK k H)P k|k-1 Formula 7

[0107] in, is the predicted state at time k; P k|k-1 is the prediction covariance matrix at time k; Kk is the Kalman gain; zk is the observation at time k; H is the observation matrix H = [1 0]; Q is the process noise covariance matrix R is the observation noise covariance matrix, default is 0.01; I is the identity matrix P0 initial state estimation error covariance matrix

[0108] (3) For the machine learning classification stage, random forest (RF) can be used for classification and the classifier can be trained to distinguish between live and non-live signals. Random forest is an ensemble learning method that performs classification or regression tasks through a combination of multiple decision trees. Among them, the basic formula of random forest includes: the generation of each tree and the final prediction. For the generation of each tree: randomly extract samples with replacement from the training set to form a subset. At each node, a subset of features is randomly selected. The node is split using the best split criterion (the random number in the information gain is 42). For the final prediction: For classification tasks, the prediction of random forest is the majority class of the voting results of all trees. For details, please refer to the following formula 8:

[0109]

[0110] Where hi(x) is the prediction of the i-th tree for the input x; B is the number of trees, which is 50 (to ensure reproducible results).

[0111] It can be understood that after obtaining the liveness detection result corresponding to the echo data of the second signal output by the liveness detection model, data processing and transmission can be performed, that is, an embedded processor can be used to process the radar data and transmit the detection result to the vehicle's main control system through the in-vehicle network (such as CAN bus).

[0112] It is also understood that the hardware block diagram of the liveness detection system can be found in Figure 7 , Figure 7 A schematic diagram of the hardware structure of a liveness detection system disclosed in an embodiment of the present application is shown in FIG. Figure 7 It can be seen that the BLE+UWB module is used on the car ceiling light controller host OHC, and the UWB module antenna is designed in 1T2R mode.

[0113] It is also understood that the software block diagram of the liveness detection system can be found in Figure 8 , Figure 8 A software structure diagram of a liveness detection system disclosed in an embodiment of the present application is shown in FIG. Figure 8 It can be seen that the MCU can obtain Bluetooth RSSI data and radar data, and perform data filtering and data processing on the Bluetooth RSSI data and radar data. It can perform liveness detection based on the filtered and processed data, and communicate with the in-vehicle network based on the liveness detection results, so that the vehicle main control system can perform corresponding control.

[0114] In order to facilitate the understanding of the embodiments of the present application, Figure 3 , Figure 7 and Figure 8, Let's take a specific example to illustrate: when the vehicle is locked, the integrated Bluetooth RSSI receiving module can be started through the ceiling light controller, and the broadcast data of the Bluetooth sensor can be periodically scanned through the Bluetooth RSSI receiving module. After determining that there is a living body, the UWB radar module is started, and the received radar signal is processed to extract characteristic parameters (such as breathing rate, heart rate), and the presence of personnel in the car and the status of the personnel are determined based on the characteristic parameters. The detection results are transmitted to the vehicle's main control system through the in-vehicle network, so that the main control system triggers corresponding responses based on the detection results, such as alarms, air conditioning adjustments, etc.

[0115] Before inputting the echo data of the second signal into the pre-trained liveness detection model, an echo data sample of the second signal sample can be obtained first, the echo data sample of the second signal sample is annotated with a liveness detection recognition result, the second signal sample is a signal emitted by a transmitter of a radar device, and the echo data sample of the second signal sample is the echo data of the second signal sample emitted by the transmitter of the radar device received by a receiver of the radar device, and then the echo data sample of the second signal sample is input into the liveness detection model, and the liveness detection model performs liveness detection of the target environment factors on the echo data sample of the second signal sample to obtain a liveness detection result corresponding to the echo data sample of the second signal sample, and finally, when the loss between the predicted liveness detection result and the annotated liveness detection result meets the convergence condition, a trained liveness detection model is obtained.

[0116] It is worth mentioning that using the second signal sample to train liveness detection in the target environment of the target environmental factor can establish a more sophisticated liveness detection model, so that the trained liveness detection model can adapt to new environments and challenges.

[0117] To verify the beneficial effects of the embodiments of the present application, a specific example is given below for illustration:

[0118] In this example, the device configuration can be to use a certain model of car ceiling light controller, integrate Bluetooth RSSI reception and UWB radar module, and connect to the vehicle main control system through the CAN bus. The test process can be to test in stationary and driving states respectively, simulating different environments and personnel states (such as sitting, lying, breathing changes), and the ceiling light controller periodically sends UWB pulses and receives and processes echo signals. For actual test data, please refer to Fig. 9 , Fig.10 , Fig.11 , Fig.12 and Fig.13 , Fig. 9 Schematic diagram of a moving average filtering of Bluetooth RSSI data disclosed in an embodiment of the present application, Fig.10 is a schematic diagram of a CWT of a respiratory signal disclosed in an embodiment of the present application, Fig.11 is a schematic diagram of HHT of a respiratory signal disclosed in an embodiment of the present application, Fig.12 A schematic diagram of Kalman filter data fusion disclosed in an embodiment of the present application is shown in FIG. Fig.13 A schematic diagram of a Kalman filter estimation result disclosed in an embodiment of the present application, comprising Fig. 9 , 10 , 11, 12 and 13, it can be seen that the normal breathing rate of a human body is usually 12 to 20 times per minute, and the frequency is approximately in the range of 0.2Hz to 0.4Hz. The actual and predicted results of RF classification are the same. Therefore, the test results obtained are that the embodiment of the present application can accurately detect the presence and status of people in the car, with a false alarm rate of less than 1%, and the detection effect remains stable in different environments (day, night, temperature changes, etc.).

[0119] It can be understood that in the embodiments of the present application, in addition to the method of first detecting Bluetooth data and then detecting radar data as described above, the method of liveness detection can also be to first detect radar data and then detect Bluetooth data. The specific method of liveness detection is not limited here.

[0120] In this embodiment, it is possible to determine whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal. If there is, the transmitter of the radar device is controlled to transmit a second signal, and the echo data of the second signal received by the receiver of the radar device is obtained, and the liveness detection model performs liveness detection of the target environment of the target environmental factor on the echo data of the second signal. The Bluetooth RSSI data (the signal strength indication of the first signal) can provide real-time signal strength change information, and the UWB radar data (the echo data of the second signal) can provide high-precision location information and motion detection. By combining the Bluetooth RSSI data and the UWB radar data, the accuracy of liveness detection is high. Further, by monitoring and considering the target environmental factors (such as temperature, humidity, and light, etc.), the liveness detection strategy can be adjusted according to the actual situation, the influence of environmental factors on the accuracy of liveness detection can be reduced, the dependence on environmental conditions can be reduced, the robustness and stability of the system can be improved, and the reliability of liveness detection can be high. Furthermore, it is possible to first determine whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal. If there is, the transmitter of the radar device is controlled to transmit the second signal, and the echo data of the second signal received by the receiver of the radar device is obtained, and the echo data of the second signal is input into the pre-trained liveness detection model, and the liveness detection model is used to perform liveness detection. Therefore, the effect of taking into account both high liveness detection accuracy and low system loss can be achieved. Secondly, the Bluetooth receiving device and the radar device can be integrated with the ceiling light controller to reduce the installation of additional equipment. This integrated design method can save space and cost. Furthermore, it is possible to first determine whether there is a living body in the target environment based on the echo data of the second signal after the second filtering process. If there is, the echo data of the second signal after the second filtering process is input into the liveness detection model to perform liveness detection through the liveness detection model. Therefore, the effect of taking into account both high liveness detection accuracy and low system loss can be achieved. Finally, it is possible to avoid using detection methods such as cameras that may infringe on privacy, protect privacy, and have strong environmental adaptability.

[0121] The above describes the liveness detection method in the embodiment of the present application. The following describes the liveness detection device in the embodiment of the present application. Fig.14 , an embodiment of the living body detection device in the embodiment of the present application includes:

[0122] The acquisition unit 1401 is used to control the Bluetooth transmitting device to periodically transmit the first signal if a trigger signal for liveness detection is detected, and to obtain a signal strength indication of the first signal received by the Bluetooth receiving device;

[0123] A determining unit 1402, configured to determine whether there is a living body in a target environment of a target environmental factor based on a distribution of a signal strength indication of the first signal, where the target environmental factor includes at least one of temperature, humidity and light;

[0124] The acquisition unit 1401 is further configured to, if present, control the transmitter of the radar device to transmit the second signal, and acquire echo data of the second signal received by the receiver of the radar device;

[0125] The model detection unit 1403 is used to input the echo data of the second signal into a pre-trained liveness detection model, and the liveness detection model performs liveness detection of the target environment factors on the echo data of the second signal to obtain a liveness detection result corresponding to the echo data of the second signal.

[0126] In the embodiment of the present application, it is possible to determine whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal. If there is, the transmitter of the radar device is controlled to transmit a second signal, and the echo data of the second signal received by the receiver of the radar device is obtained, and the liveness detection model performs liveness detection of the target environment of the target environmental factor on the echo data of the second signal. The Bluetooth RSSI data (the signal strength indication of the first signal) can provide real-time signal strength change information, and the UWB radar data (the echo data of the second signal) can provide high-precision location information and motion detection. By combining the Bluetooth RSSI data and the UWB radar data, the accuracy of liveness detection is high. Further, by monitoring and considering the target environmental factors (such as temperature, humidity, and light, etc.), the liveness detection strategy can be adjusted according to the actual situation, the influence of environmental factors on the accuracy of liveness detection can be reduced, the dependence on environmental conditions can be reduced, the robustness and stability of the system can be improved, and the reliability of liveness detection can be high. Furthermore, it is possible to first determine whether there is a living organism in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal. If so, the transmitter of the radar device is controlled to transmit a second signal, and the echo data of the second signal received by the receiver of the radar device is obtained, and the echo data of the second signal is input into a pre-trained liveness detection model, and the liveness detection is performed by the liveness detection model. Therefore, the effect of both high liveness detection accuracy and low system loss can be achieved.

[0127] The following is a detailed description of the liveness detection device in the embodiment of the present application. Fig.15 Another embodiment of the liveness detection in the embodiment of the present application includes:

[0128] The acquisition unit 1501 is configured to control the Bluetooth transmitting device to periodically transmit a first signal if a trigger signal for liveness detection is detected, and to obtain a signal strength indication of the first signal received by the Bluetooth receiving device;

[0129] A determining unit 1502 is configured to determine whether there is a living body in a target environment of a target environmental factor based on a distribution of the signal strength indication of the first signal, wherein the target environmental factor includes at least one of temperature, humidity and light;

[0130] The acquisition unit 1501 is further configured to, if present, control the transmitter of the radar device to transmit a second signal, and acquire echo data of the second signal received by the receiver of the radar device;

[0131] The model detection unit 1503 is used to input the echo data of the second signal into a pre-trained liveness detection model, and the liveness detection model performs liveness detection of the target environment of the target environmental factors on the echo data of the second signal to obtain a liveness detection result corresponding to the echo data of the second signal.

[0132] The determination unit 1502 is specifically used to determine whether the fluctuation range of the signal strength indication of the first signal is within the preset living body fluctuation range corresponding to the target environment of the target environment factor, and determine whether there is a living body in the target environment of the target environment factor based on the judgment result; and / or determine whether the intensity value range of the signal strength indication of the first signal is within the preset living body intensity value range corresponding to the target environment of the target environment factor, and determine whether there is a living body in the target environment of the target environment factor based on the judgment result.

[0133] The model detection unit 1503 is specifically used to determine whether the echo data of the second signal has a preset liveness feature of the target environment of the target environmental factor. If so, it is determined that the liveness detection result corresponding to the echo data of the second signal is a liveness in the target environment with the target environmental factor. If not, it is determined that the liveness detection result corresponding to the echo data of the second signal is a liveness in the target environment without the target environmental factor.

[0134] The acquisition unit 1501 is specifically configured to acquire a signal strength indication of an initial signal, and determine the first signal sent by the Bluetooth sending module in the initial signal, wherein the initial signal includes a signal transmitted by at least one Bluetooth sending device;

[0135] The determination unit 1502 is specifically configured to perform a first filtering process on the first signal to obtain a first signal after the first filtering process, and determine whether there is a living body in the target environment of the target environmental factor based on the distribution of the signal strength indication of the first signal after the filtering process.

[0136] The model detection unit 1503 is specifically used to perform a second filtering process on the echo data of the second signal to obtain the echo data of the second signal after the second filtering process. If it is determined that there is a living body in the target environment based on the echo data of the second signal after the second filtering process, the echo data of the second signal after the second filtering process is input into the living body detection model. The living body detection model performs living body detection of the target environment factor on the echo data of the second signal after the second filtering process to obtain a living body detection result corresponding to the echo data of the second signal after the second filtering process.

[0137] The determination unit 1502 is specifically used to extract multi-dimensional features of the echo data of the second signal to obtain the multi-dimensional features of the echo data of the second signal, and to fuse the multi-dimensional features of the echo data of the second signal to obtain fused features, where the fused features are the echo data of the second signal after the second filtering process.

[0138] The determination unit 1502 is specifically used to perform time-frequency analysis on the echo data of the second signal according to a continuous wavelet transform algorithm to obtain time-frequency dimensional characteristics of the echo data of the second signal, and the time-frequency dimensional characteristics of the echo data of the second signal are multi-dimensional characteristics of the echo data of the second signal; and / or perform amplitude and phase analysis on the echo data of the second signal according to a Hilbert transform algorithm to obtain amplitude and phase dimensional characteristics of the echo data of the second signal, and the amplitude and phase dimensional characteristics of the echo data of the second signal are multi-dimensional characteristics of the echo data of the second signal.

[0139] The living body detection device also includes:

[0140] The obtaining unit 1504 is used to obtain an echo data sample of a second signal sample, where the echo data sample of the second signal sample is marked with a liveness detection and recognition result, where the second signal sample is a signal transmitted by a transmitter of the radar device, and the echo data sample of the second signal sample is echo data of the second signal sample transmitted by the transmitter of the radar device and received by a receiver of the radar device;

[0141] The model detection unit 1503 is further configured to input the echo data sample of the second signal sample into a liveness detection model, and the liveness detection model performs liveness detection of the target environment of the target environment factor on the echo data sample of the second signal sample to obtain a liveness detection result corresponding to the echo data sample of the second signal sample;

[0142] The calculation unit 1505 is used to obtain a trained liveness detection model when the loss between the predicted liveness detection result and the labeled liveness detection result meets a convergence condition.

[0143] In this embodiment, each unit in the living body detection device performs the above Figure 2 The operation of the liveness detection method in the illustrated embodiment will not be described in detail here.

[0144] See below Fig.16 Another embodiment of the liveness detection device 1600 in the embodiment of the present application includes:

[0145] CPU 1601, memory 1605, input / output interface 1604, wired or wireless network interface 1603 and power supply 1602;

[0146] The memory 1605 is a temporary storage memory or a permanent storage memory;

[0147] The CPU 1601 is configured to communicate with the memory 1605 and execute the instructions in the memory 1605 to perform the aforementioned Figure 2 The method in the illustrated embodiment.

[0148] The embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium includes instructions, when the instructions are executed on a computer, the computer executes the aforementioned Figure 2 The method in the illustrated embodiment.

[0149] The present application also provides a computer program product including instructions, which, when executed on a computer, enables the computer to execute the aforementioned Figure 2 The method in the illustrated embodiment.

[0150] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0151] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0152] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units 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 an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0153] 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.

[0154] 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.

[0155] 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 storage medium. Based on this understanding, the technical solution of the present application is essentially 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, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk and other media that can store program code.

Claims

1. A method for detecting a living body, characterized in that: include: If a trigger signal for liveness detection is detected, the Bluetooth transmitting device is controlled to periodically transmit a first signal, and a signal strength indication of the first signal received by the Bluetooth receiving device is obtained; Determine whether there is a living body in a target environment of a target environmental factor based on a distribution of the signal strength indication of the first signal, wherein the target environmental factor includes at least one of temperature, humidity and light; If so, controlling the transmitter of the radar device to transmit a second signal, and acquiring echo data of the second signal received by the receiver of the radar device; The echo data of the second signal is input into a pre-trained liveness detection model, and the liveness detection model performs liveness detection of the target environment of the target environmental factors on the echo data of the second signal to obtain a liveness detection result corresponding to the echo data of the second signal output by the liveness detection model.

2. The method according to claim 1, characterized in that The determining whether there is a living body in the target environment of the target environment factor based on the distribution of the signal strength indication of the first signal comprises: Determine whether the fluctuation range of the signal strength indication of the first signal is within a preset living body fluctuation range corresponding to the target environment of the target environmental factor, and determine whether there is a living body in the target environment of the target environmental factor based on the judgment result; and / or Determine whether the intensity value range of the signal intensity indication of the first signal is within a preset living body intensity value range corresponding to the target environment of the target environment factor, and determine whether there is a living body in the target environment of the target environment factor based on the judgment result.

3. The method according to claim 1, characterized in that The performing liveness detection of the target environment of the target environmental factor on the echo data of the second signal to obtain a liveness detection result corresponding to the echo data of the second signal includes: Determine whether the echo data of the second signal has a preset living feature of the target environment of the target environmental factor; If so, determining that the living body detection result corresponding to the echo data of the second signal is a living body in the target environment where the target environmental factor exists; If not, it is determined that the living body detection result corresponding to the echo data of the second signal is a living body in the target environment where the target environmental factor does not exist.

4. The method according to claim 1, characterized in that: The obtaining the signal strength indication of the first signal received by the Bluetooth receiving device includes: Acquire a signal strength indication of an initial signal, and determine the first signal sent by the Bluetooth sending module in the initial signal, wherein the initial signal includes a signal transmitted by at least one Bluetooth sending device; The determining whether there is a living body in the target environment of the target environment factor based on the distribution of the signal strength indication of the first signal comprises: Performing a first filtering process on the first signal to obtain a first signal after the first filtering process; It is determined whether there is a living body in the target environment of the target environment factor based on the distribution of the signal strength indication of the first signal after the filtering process.

5. The method according to claim 1, characterized in that The step of inputting the echo data of the second signal into a pre-trained liveness detection model, performing liveness detection of the target environment of the target environment factor on the echo data of the second signal by the liveness detection model, and obtaining a liveness detection result corresponding to the echo data of the second signal includes: Performing a second filtering process on the echo data of the second signal to obtain the echo data of the second signal after the second filtering process; If it is determined that there is a living organism in the target environment based on the echo data of the second signal after the second filtering processing, the echo data of the second signal after the second filtering processing is input into the living organism detection model, and the living organism detection model performs living organism detection of the target environment factors on the echo data of the second signal after the second filtering processing, and obtains the living organism detection result corresponding to the echo data of the second signal after the second filtering processing.

6. The method according to claim 5, characterized in that The performing a second filtering process on the echo data of the second signal to obtain the echo data of the second signal after the second filtering process includes: Extracting multi-dimensional features from the echo data of the second signal to obtain the multi-dimensional features of the echo data of the second signal; The multi-dimensional features of the echo data of the second signal are fused to obtain a fused feature, where the fused feature is the echo data of the second signal after the second filtering process.

7. The method according to claim 6, characterized in that The extracting of multi-dimensional features from the echo data of the second signal to obtain the multi-dimensional features of the echo data of the second signal includes: Performing time-frequency analysis on the echo data of the second signal according to a continuous wavelet transform algorithm to obtain time-frequency dimensional features of the echo data of the second signal, wherein the time-frequency dimensional features of the echo data of the second signal are multi-dimensional features of the echo data of the second signal; and / or The amplitude and phase analysis of the echo data of the second signal is performed according to the Hilbert transform algorithm to obtain the amplitude and phase dimension characteristics of the echo data of the second signal. The amplitude and phase dimension characteristics of the echo data of the second signal are the multi-dimensional characteristics of the echo data of the second signal.

8. The method according to claim 1, characterized in that: Before inputting the echo data of the second signal into a pre-trained living body detection model, the method further includes: Obtain an echo data sample of a second signal sample, wherein the echo data sample of the second signal sample is annotated with a liveness detection and recognition result, wherein the second signal sample is a signal transmitted by a transmitter of the radar device, and the echo data sample of the second signal sample is echo data of the second signal sample transmitted by the transmitter of the radar device and received by a receiver of the radar device; Inputting the echo data sample of the second signal sample into a liveness detection model, and performing liveness detection of the target environment of the target environment factor on the echo data sample of the second signal sample by the liveness detection model to obtain a liveness detection result corresponding to the echo data sample of the second signal sample; When the loss between the predicted liveness detection result and the labeled liveness detection result meets the convergence condition, a trained liveness detection model is obtained.

9. A living body detection device, characterized in that: include: an acquisition unit, configured to control the Bluetooth transmitting device to periodically transmit a first signal if a trigger signal for liveness detection is detected, and to acquire a signal strength indication of the first signal received by the Bluetooth receiving device; A determination unit, configured to determine whether there is a living body in a target environment of a target environmental factor based on a distribution of a signal strength indication of the first signal, wherein the target environmental factor includes at least one of temperature, humidity and light; The acquisition unit is further configured to, if present, control the transmitter of the radar device to transmit a second signal, and acquire echo data of the second signal received by the receiver of the radar device; A model detection unit is used to input the echo data of the second signal into a pre-trained liveness detection model, and the liveness detection model performs liveness detection of the target environment of the target environmental factors on the echo data of the second signal to obtain a liveness detection result corresponding to the echo data of the second signal.

10. A living body detection device, characterized in that: include: CPU and memory; The memory is a transient storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instructions in the memory to perform the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium comprises instructions, and when the instructions are executed on a computer, the computer is caused to perform the method according to any one of claims 1 to 8.

Images