Living body detection method, device and equipment, vehicle and storage medium

Abstract

The invention discloses a living body detection method, a living body detection device, electronic equipment, a vehicle and a computer readable storage medium. The in-vivo detection method is used for a vehicle, and specifically comprises the following steps: in response to an ultra-wideband pulse signal generated in a vehicle cabin, obtaining original channel pulse response data fed back by an internal environment of the vehicle cabin; performing signal preprocessing on the original channel pulse response data to determine channel pulse data so as to improve the accuracy of a living body detection signal; and determining a living body detection signal according to the channel pulse data so as to detect whether a living body exists in the vehicle cabin or not. According to the method, the CIR data in the internal environment of the vehicle cabin can be obtained by receiving and transmitting UWB signals in the vehicle cabin, signal preprocessing is firstly performed before living body detection judgment is performed based on the CIR data so as to adjust the parameter performance of the obtained CIR data which can influence the quality of the CIR data, the accuracy of the CIR data is improved, and the accuracy of the CIR data is improved. And thus, the effectiveness degree and the accuracy degree of living body detection are improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent vehicle technology, and in particular to a liveness detection method, a liveness detection device, an electronic device, a vehicle, and a computer-readable storage medium. Background Art

[0002] As the number of vehicles increases, there have been many incidents of infants or pets being forgotten in cars, resulting in casualties. In order to deal with the above situation and reduce the probability of the above incidents, the current relevant technology generally adopts the installation of new equipment such as in-car environment monitoring systems or seat alarm systems, etc., which can use sound and light alarms to remind parking lot managers or car owners when infants or pets are forgotten in the vehicle cabin. However, such a solution has difficulties in modifying existing vehicles on the one hand, and defects in detection accuracy on the other hand, which may lead to false detection or missed detection, and then lead to adverse situations such as disturbing users or life safety risks of living beings in the vehicle cabin. Summary of the invention

[0003] The present application provides a liveness detection method, a liveness detection device, an electronic device, a vehicle, and a computer-readable storage medium.

[0004] The liveness detection method involved in the embodiment of the present application is used in a vehicle, and the method comprises the following steps:

[0005] In response to the ultra-wideband pulse signal generated into the interior of the vehicle cabin, obtaining original channel impulse response data fed back from the interior environment of the vehicle cabin;

[0006] Performing signal preprocessing on the original channel impulse response data to determine channel impulse data to improve the accuracy of the living body detection signal;

[0007] A living body detection signal is determined according to the channel pulse data to detect whether a living body exists in the vehicle cabin.

[0008] In this way, the present application can utilize the transmission and reception of UWB signals inside the vehicle cabin to obtain CIR data in the vehicle cabin environment, and first perform signal preprocessing before performing liveness detection judgment based on the CIR data to adjust the acquired CIR data's effective signal ratio, signal-to-noise ratio, burr ratio and other parameters that can affect the CIR data quality, thereby improving the accuracy of the CIR data, and then improving the accuracy of the liveness detection signal as well as the effectiveness and accuracy of liveness detection.

[0009] In some implementations, the performing signal preprocessing on the original channel impulse response data to determine the channel impulse data includes:

[0010] Pre-analysis processing is performed on the original channel impulse response data to determine first channel impulse response data so as to sort out effective signals in the original channel impulse response data.

[0011] In this way, the present application can pre-parse the acquired original CIR data in advance, thereby sorting out the valid parts of the acquired signal and reducing the interference caused by the invalid parts.

[0012] In some implementations, the performing signal preprocessing on the original channel impulse response data to determine the channel impulse data further includes:

[0013] A packet adaptation process is performed on the first channel impulse response data to determine the channel impulse data so that the channel impulse data can be recognized by the vehicle.

[0014] In this way, the present application can further package and adapt the sorted effective signals, so that the effective CIR data can be recognized by various vehicle-mounted systems of various vehicles, thereby expanding the scope of application of the above method.

[0015] In some implementations, the performing signal preprocessing on the original channel impulse response data to determine the channel impulse data further includes:

[0016] An impulse integration process is performed on the first channel impulse response data to determine second channel impulse data so as to amplify a non-noise portion in the first channel impulse response data.

[0017] In this way, the present application can further perform pulse accumulation processing on the basis of sorting out the effective part of the acquired CIR data to improve the signal-to-noise ratio of the effective CIR data.

[0018] In some implementations, the performing signal preprocessing on the original channel impulse response data to determine the channel impulse data further includes:

[0019] An outlier elimination process is performed on the second channel pulse data to determine the third channel pulse data.

[0020] In this way, the present application can further eliminate outliers in the signal on the basis of improving the signal-to-noise ratio, thereby further improving the accuracy of the CIR data.

[0021] In some implementations, the performing signal preprocessing on the original channel impulse response data to determine the channel impulse data further includes:

[0022] The third channel pulse data is filtered to determine the channel pulse data so as to remove burrs in the channel pulse data.

[0023] In this way, the present application can also use filtering processing to remove burrs in CIR data when abnormal values ​​have been removed, thereby improving the accuracy of CIR data.

[0024] In some embodiments, determining the living body detection signal according to the channel pulse data includes:

[0025] Performing Fourier transform according to the channel pulse data to determine a response frequency domain signal;

[0026] The living body detection signal is statistically determined according to the response frequency domain signal and preset scene feature parameters.

[0027] In this way, the present application can perform frequency domain processing and statistical processing based on the CIR data determined by preprocessing, perform data statistics based on preset scene feature parameters of the presence and absence of living things in the vehicle cabin, and finally determine the liveness detection result in the form of a liveness detection signal.

[0028] In certain embodiments, the method further comprises:

[0029] The living body detection signal is sent to a matching smart device to inform the user of the presence of a living body in the vehicle cabin.

[0030] In this way, the present application can also send the liveness detection result to the smart device through the communication relationship between the smart devices used in conjunction with the vehicle when the liveness detection result is determined, thereby providing timely and accurate reminders to users regarding the problem of living things left behind in the vehicle cabin, thereby reducing the probability of life risks to living things in the vehicle cabin due to forgetfulness.

[0031] The liveness detection device in the embodiment of the present application is used in a vehicle and includes:

[0032] A signal receiving module, configured to obtain original channel impulse response data fed back from the internal environment of the vehicle cabin in response to an ultra-wideband pulse signal generated to the interior of the vehicle cabin;

[0033] A signal processing module, used for performing signal preprocessing on the original channel impulse response data to determine channel impulse data so as to improve the accuracy of the living body detection signal;

[0034] The signal detection module is used to determine a living body detection signal according to the channel pulse data to detect whether there is a living body in the vehicle cabin.

[0035] In some embodiments, the living body detection device further comprises:

[0036] The signal transceiver module is used to send the living body detection signal to a matching smart device to inform the user of the presence of living bodies in the vehicle cabin.

[0037] The electronic device in the implementation mode of the present application includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the above method is implemented.

[0038] The vehicle in the embodiment of the present application includes the above-mentioned living body detection device or the above-mentioned electronic device.

[0039] The computer-readable storage medium in the embodiments of the present application stores a computer program, and when the computer program is executed by one or more processors, the above method is implemented.

[0040] Additional aspects and advantages of the embodiments of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0042] Figure 1 This is a flow chart of a liveness detection method in an embodiment of the present application;

[0043] Figure 2 This is a flow chart of a liveness detection method in an embodiment of the present application;

[0044] Figure 3 This is a flow chart of a liveness detection method in an embodiment of the present application;

[0045] Figure 4 This is a flow chart of a liveness detection method in an embodiment of the present application;

[0046] Figure 5 This is a flow chart of a liveness detection method in an embodiment of the present application;

[0047] Figure 6 This is a flow chart of a liveness detection method in an embodiment of the present application;

[0048] Figure 7 Schematic diagram of the flow of the liveness detection method in the embodiment of the present application. DETAILED DESCRIPTION

[0049] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions from beginning to end. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the embodiments of the present application, and cannot be understood as limiting the embodiments of the present application.

[0050] See also Figure 1 The liveness detection method in the embodiment of the present application is used for a vehicle and specifically comprises the following steps:

[0051] 01: In response to the ultra-wideband pulse signal generated inside the vehicle cabin, the original channel impulse response data of the vehicle cabin internal environment feedback is obtained;

[0052] 02: Perform signal preprocessing on the original channel impulse response data to determine the channel impulse data to improve the accuracy of the liveness detection signal;

[0053] 03: Determine a liveness detection signal based on the channel pulse data to detect whether there is a live person in the vehicle cabin.

[0054] The liveness detection device in the embodiment of the present application can implement the liveness detection method described above. Specifically, the liveness detection device includes a signal receiving module, a signal processing module and a signal detection module. The signal receiving module is used to respond to the ultra-wideband pulse signal generated inside the vehicle cabin to obtain the original channel pulse response data fed back by the internal environment of the vehicle cabin, the signal processing module is used to perform signal preprocessing on the original channel pulse response data and determine the channel pulse data to improve the accuracy of the liveness detection signal, and the signal detection module is used to determine the liveness detection signal based on the channel pulse data to detect whether there is a live body in the vehicle cabin.

[0055] The electronic device in the embodiment of the present application can implement the above-mentioned liveness detection method. Specifically, the electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor is used to respond to the ultra-wideband pulse signal generated inside the vehicle cabin, obtain the original channel pulse response data fed back by the internal environment of the vehicle cabin, and perform signal preprocessing on the original channel pulse response data to determine the channel pulse data to improve the accuracy of the liveness detection signal, and to determine the liveness detection signal according to the channel pulse data to detect whether there is a live body in the vehicle cabin.

[0056] Specifically, in order to solve the problems existing in the above-mentioned background technology, the present application proposes a method for detecting living bodies, which mainly adopts the emission of ultra-wideband (hereinafter referred to as UWB, Ultra Wide Band) pulse signals inside the vehicle cabin and the reception of echoes to detect whether there are living bodies inside the vehicle cabin. The above scheme can use the Doppler effect during the propagation of UWB pulse signals to detect changes in the surrounding electromagnetic environment. When a microwave beam in the form of a pulse irradiates the human body or animal body, due to the existence of life movements such as breathing and gastrointestinal peristalsis, the repetition period and phase of the echo pulse sequence scattered by the human body will change compared with the pulse sequence at the time of emission, and the repetition period of the echo pulse signal is related to the speed and frequency of the life activities of the living body itself. If the pulse sequence and phase are mediated, integrated, amplified, filtered, etc. and sent to a computer for data processing and analysis, parameters related to the vital signs of the human body being measured are obtained, thereby realizing the detection of living bodies.

[0057] The basic principle of using UWB pulse signals is that the UWB device transmits UWB pulse signals to the inside of the vehicle cabin, receives the echo of the pulse signal after being reflected or scattered by obstacles, and judges whether there are living things or objects near the UWB device by analyzing the echo disturbance. Specifically, the UWB device detects surrounding objects and their movements through the received channel impulse response data (hereinafter referred to as CIR, Channel Impulse Response). When the surrounding objects move, the phase of the corresponding CIR will change due to the Doppler effect. The reflected frequency-shifted signal is amplified by the active filter in the Doppler microwave sensor controller, and is sent to the A / D port for sampling after filtering. The A / D value of the signal collected at the current moment of the object's movement will be significantly different from the preset threshold. Therefore, the movement state of the living body can be estimated by analyzing the change in CIR data, thereby achieving the purpose of detecting the living body.

[0058] Therefore, based on the above principles, the liveness detection method in the embodiments of the present application can be implemented in the following manner:

[0059] First, based on the UWB radar or other types of UWB devices installed in the vehicle, a UWB pulse signal is transmitted to the interior of the vehicle cabin. At the same time, the UWB device is ready to receive the echo signal returned from the interior of the vehicle cabin based on the above pulse signal. When the UWB device obtains the above echo signal, it will directly generate the corresponding original CIR data.

[0060] Next, the acquired raw CIR data is preprocessed. The purpose of preprocessing is to extract and organize the effective data with high reference value in the raw CIR data to improve the accuracy and efficiency of liveness detection based on CIR data. In addition, preprocessing can also include removing some data that cause greater interference to the detection process, such as noise, glitches, and abnormal data, to further improve the accuracy of the liveness detection process.

[0061] Finally, data analysis is performed on the pre-processed CIR data, and the detection and judgment results of life in the vehicle cabin are determined based on the overall and local changes in the CIR data, thus completing the entire liveness detection process.

[0062] In this way, the present application can utilize the transmission and reception of UWB signals inside the vehicle cabin to obtain CIR data in the vehicle cabin environment, and first perform signal preprocessing before performing liveness detection judgment based on the CIR data to adjust the acquired CIR data's effective signal ratio, signal-to-noise ratio, burr ratio and other parameters that can affect the CIR data quality, thereby improving the accuracy of the CIR data, and then improving the accuracy of the liveness detection signal as well as the effectiveness and accuracy of liveness detection.

[0063] See also Figure 2 In some embodiments, step 02 includes:

[0064] 021: Perform pre-analysis processing on the original channel impulse response data to determine first channel impulse response data, so as to sort out valid signals in the original channel impulse response data.

[0065] In some implementations, the signal processing module is further configured to perform pre-analysis processing on the original channel impulse response data to determine first channel impulse response data so as to sort out valid signals in the original channel impulse response data.

[0066] In some implementations, the processor is further configured to perform a pre-analysis process on the original channel impulse response data to determine first channel impulse response data, so as to sort out valid signals in the original channel impulse response data.

[0067] Specifically, on the basis of the above implementation mode, on the basis that the UWB device has obtained the original CIR data, since the emission of the UWB pulse signal and the acquisition of the echo may be interfered by the electromagnetic environment of the outside world or the inside of the vehicle cabin, which may cause some data to be obviously distorted, in order to improve the accuracy and recognition efficiency when using CIR data for liveness recognition, it is generally necessary to remove some CIR data that has no reference value for the liveness recognition process in advance. Therefore, in some examples, when the UWB device has obtained the original CIR data, the original CIR data is first parsed and read, and the original CIR data is sorted and screened based on the preset valid CIR data range, and the valid CIR data that meets the valid CIR data range is sorted out and retained in the form of the first channel pulse response data, and some CIR data that does not meet the valid CIR data range is shielded or discarded. In the subsequent liveness recognition, the first channel pulse response data is used as the recognition basis, and the above-mentioned part of the CIR data that does not conform to the valid CIR data range will be ignored. In this way, on the one hand, the accuracy of liveness recognition based on CIR data is improved and interference caused by invalid data is avoided. On the other hand, the total amount of data required for recognition and judgment in the liveness recognition process is reduced, thereby improving the efficiency of the liveness recognition process.

[0068] In this way, the present application can pre-parse the acquired original CIR data in advance, thereby sorting out the valid parts of the acquired signal and reducing the interference caused by the invalid parts.

[0069] See also Figure 3 In some embodiments, step 02 further includes:

[0070] 022: Perform packaging adaptation processing on the first channel impulse response data to determine the channel impulse data so that the channel impulse data can be recognized by the vehicle.

[0071] In some embodiments, the signal processing module is further used to perform packaging adaptation processing on the first channel impulse response data to determine the channel impulse data so that the channel impulse data can be recognized by the vehicle.

[0072] In some embodiments, the processor is further configured to perform a packing adaptation process on the first channel impulse response data to determine the channel impulse data so that the channel impulse data can be recognized by the vehicle.

[0073] Specifically, based on the above implementation, due to the different types and performances of processors in different vehicles, the data formats that can be processed are also different. The first channel impulse response data obtained through the data screening and sorting process in the above example may not be directly supported by the vehicle's processor in terms of format and can be fully processed, that is, the types of vehicles that can directly process the first channel impulse response data may be limited. Moreover, since the original CIR data is data that changes continuously with the changes in the space in the vehicle cabin, after parsing and screening based on the method in the above example, the data may become discontinuous data, and this discontinuity also interferes with subsequent recognition.

[0074] Therefore, in order to enable more types of vehicles and more types of processors to fully process the first channel impulse response data, illustratively, when valid first channel impulse response data has been screened out through sorting and analysis, the discontinuous first channel impulse response data is first packaged, so that the sorted valid data is integrated into a complete data packet to improve the subsequent data processing efficiency. Secondly, while packaging, the first channel impulse response data is also mathematically adapted, so that most processors can directly process the data, thereby expanding the scope of application of the above-mentioned liveness detection method in vehicles currently on the market.

[0075] In this way, the present application can further package and adapt the sorted effective signals, so that the effective CIR data can be recognized by various vehicle-mounted systems of various vehicles, thereby expanding the scope of application of the above method.

[0076] See also Figure 4 In some embodiments, step 02 further includes:

[0077] 023: Perform pulse accumulation processing on the first channel impulse response data to determine the second channel impulse data so as to amplify the non-noise portion in the first channel impulse response data.

[0078] In some embodiments, the signal processing module is further configured to perform pulse accumulation processing on the first channel impulse response data to determine the second channel impulse data so as to amplify a non-noise portion in the first channel impulse response data.

[0079] In some embodiments, the processor is further configured to perform a pulse accumulation process on the first channel impulse response data to determine the second channel impulse data so as to amplify a non-noise portion in the first channel impulse response data.

[0080] Specifically, based on the above implementation, when using UWB pulse signals to detect targets at a long distance, since the target echo signal is weaker than in the average state and the signal amplitude is very small, its signal strength is difficult to distinguish from the noise signal, which can easily lead to the signal-to-noise ratio of the CIR data in the received echo signal being too low, resulting in the inability to accurately identify the target living being or even missed detection, which may lead to serious adverse consequences.

[0081] In order to avoid the above situation, in some examples, when the effective first channel impulse response data has been screened out through sorting and analysis, pulse accumulation is performed on the first channel impulse response data, and the above second channel impulse response data is obtained after pulse accumulation. The main purpose of the above pulse accumulation is to amplify the non-noise part in the first channel impulse response data. During the pulse accumulation process, the non-noise part in the first channel impulse response data will accumulate regularly due to the superposition of signal amplitudes, but the change of the noise part is not as large as the superposition amplitude of the signal amplitude, so the change of the noise part is smaller than that of the non-noise part. In this way, the pulse accumulation process can effectively amplify the non-noise part without amplifying the noise part. After repeating the pulse accumulation process for many times, the signal-to-noise ratio of the CIR data can be effectively improved, thereby highlighting the effective part of the CIR data that may detect the corresponding signal of the living body, and reducing the interference of the noise signal on the living body detection process.

[0082] Furthermore, in some examples, the above-mentioned pulse accumulation process can generally be implemented by coherent accumulation. Coherent accumulation refers to the simultaneous complex accumulation of the phase and amplitude of multiple groups of CIR data. Generally speaking, the phase of the non-noise part of the living signal that can be detected in the echo signal is coherent, and its phase is regular, while the phase of the noise signal is non-coherent. When coherent accumulation is performed, when the vectors corresponding to the noise data are added, due to the randomness of the phase, in most cases most of the noise vectors will be offset by addition. After multiple coherent accumulation cycles are completed, the sum of all noise vectors is not much different from a single noise pulse.

[0083] For example, the first channel impulse response data includes the noise part ω and the complex envelope: The non-noise part of , where A is the amplitude of the non-noise part, e is a natural constant, i is an imaginary unit, is the phase of the non-noise part. Let the noise power be σ 2 , then we know that the original signal-to-noise ratio Y1=A 2 / σ 2If the echo response is the same and the noise samples are independent at each observation, then the observation value of the echo signal will generate a new observation value Z after accumulation. The summation process of the new observation value Z to the complex signal sample is the above-mentioned coherent accumulation process:

[0084]

[0085] Among them, N-1 represents the number of coherent accumulations, ω[n] is the mean value of the noise part in a single echo and obeys statistical distribution. Then the effective signal power obtained after multiple coherent accumulations should be N 2 A 2 , the total noise power is Nσ 2 , we can calculate the accumulated signal-to-noise ratio Y2=N 2 A 2 / Nσ 2 , that is, the accumulated signal-to-noise ratio Y2 is N times larger than the original signal-to-noise ratio Y1.

[0086] In addition, in addition to the above-mentioned coherent accumulation method, illustratively, a non-coherent accumulation method can also be used, that is, only the amplitude of multiple groups of CIR data is accumulated. In simple terms, the amplitude data of the CIR data is averaged, thereby reducing the energy average deviation of the noise signal after the non-coherent accumulation process. After the non-coherent accumulation process is performed in multiple cycles, the phenomenon of loss of the non-noise part caused by the noise part can be greatly reduced, thereby amplifying the non-noise part to improve the signal-to-noise ratio. Although its effect of improving the signal-to-noise ratio is slightly weaker than that of coherent accumulation, it can be used as an alternative solution to improve the signal-to-noise ratio.

[0087] In this way, the present application can further perform pulse accumulation processing on the basis of sorting out the effective part of the acquired CIR data to improve the signal-to-noise ratio of the effective CIR data.

[0088] See also Figure 5 In some embodiments, step 02 further includes:

[0089] 024: Perform outlier elimination processing on the second channel pulse data to determine the third channel pulse data.

[0090] In some implementations, the signal processing module is further configured to perform outlier elimination processing on the second channel pulse data to determine the third channel pulse data.

[0091] In some implementations, the processor is further configured to perform outlier elimination processing on the second channel pulse data to determine the third channel pulse data.

[0092] Specifically, on the basis of the above-mentioned implementation mode, after the pulse accumulation process is completed, the obtained second channel pulse data often has abnormal values ​​with large deviations compared to the average value range. These abnormal values ​​will have a greater negative impact on the subsequent recognition process, and may cause misidentification or missed recognition. In order to avoid misidentification or missed recognition, in some examples, when the second channel pulse data is obtained through the pulse accumulation process, the second channel pulse data is further subjected to abnormal value elimination processing. Exemplarily, the above-mentioned abnormal value elimination processing process generally uses the center value of the second channel pulse data. After determining the second channel pulse data, the groups of data therein are arranged in order of size and the center value is found, and all values ​​that exceed the preset value range around the center value are determined as abnormal values. Finally, the difference between the abnormal value and the above-mentioned center value is calculated, and the absolute value of the difference is retained while its sign is kept consistent with the sign of the above-mentioned abnormal value relative to the center value. The final value is replaced with the corresponding abnormal value, and the third channel pulse data is determined, that is, the elimination of abnormal values ​​can be achieved.

[0093] In this way, the present application can further eliminate outliers in the signal on the basis of improving the signal-to-noise ratio, thereby further improving the accuracy of the CIR data.

[0094] See also Figure 6 In some embodiments, step 02 includes:

[0095] 025: Perform filtering processing on the third channel pulse data to determine the channel pulse data so as to remove burrs in the channel pulse data.

[0096] In some embodiments, the signal processing module is further configured to perform filtering processing on the third channel pulse data to determine the channel pulse data so as to remove burrs in the channel pulse data.

[0097] In some embodiments, the processor is further configured to perform filtering processing on the third channel pulse data to determine the channel pulse data so as to remove burrs in the channel pulse data.

[0098] Specifically, on the basis of the above-mentioned implementation mode, as mentioned above, the noise control method through the pulse accumulation process, whether it is coherent accumulation or non-coherent accumulation, in principle, multiple accumulations within a limited range only roughly control the noise part within a very small range, but the noise part cannot be completely eliminated, and some tiny noise parts will continue to exist in the third channel pulse data in the form of data glitches. In order to further improve the accuracy of living body recognition, in some examples, low-pass filtering is performed on the third channel pulse data by filtering. The above-mentioned low-pass filtering process generally averages or weighted averages part of the data in the sliding window of the third channel pulse data. Similar to the above-mentioned non-coherent accumulation process, by setting different average or weighted average targets, the data glitches can be suppressed to a negligible range after multiple average processes, thereby achieving burr elimination while retaining the main features of the non-noise part of the data.

[0099] In this way, the present application can also use filtering processing to remove burrs in CIR data when abnormal values ​​have been removed, thereby improving the accuracy of CIR data.

[0100] See also Figure 7 In some embodiments, step 03 includes:

[0101] 031: Perform Fourier transform according to the channel pulse data to determine the response frequency domain signal;

[0102] 032: According to the response frequency domain signal and the preset scene characteristic parameters, the living body detection signal is statistically determined.

[0103] In some embodiments, the signal detection module is further used to perform Fourier transform based on the channel pulse data to determine the response frequency domain signal, and to statistically determine the living body detection signal based on the response frequency domain signal and preset scene feature parameters.

[0104] In some embodiments, the processor is further used to perform Fourier transform based on the channel pulse data to determine the response frequency domain signal, and to statistically determine the living body detection signal based on the response frequency domain signal and preset scene feature parameters.

[0105] Specifically, after undergoing the preprocessing process in the above implementation, the signal pulse data finally obtained is the CIR data that can be directly used to perform the liveness detection and identification process. Considering that the process of performing data statistical analysis based on CIR data to achieve liveness detection needs to be processed in the frequency domain, not the time domain, and the CIR data obtained and further preprocessed according to the above implementation is a time domain signal. Therefore, before starting to perform liveness detection, it is necessary to preferentially use Fourier transform to transform the above channel pulse data from a time domain signal to a frequency domain signal.

[0106] Next, since the scene characteristics inside the vehicle cabin are generally divided into two types: scenes where living things exist and scenes where living things do not exist, the statistical characteristics of the overall CIR data in these two scenes have different representative characteristics. Therefore, in order to improve the accuracy of liveness detection, in some examples, when performing liveness detection based on CIR data that has undergone Fourier transform and belongs to frequency domain signals, at least two scene feature parameters are generally used, one is a scene feature parameter in which there is no living thing in the vehicle cabin, and the other is a scene feature parameter in which there is a living thing in the vehicle cabin. The above two scene feature parameters are generally obtained by performing a large amount of pre-statistics based on the current known data, and are feature parameters in their respective scenes, which can respectively characterize the different characteristics of whether there is a living thing in the vehicle cabin. Using the above two scene feature parameters as the judgment criteria, the CIR data that has been converted into frequency domain signals is statistically analyzed, so that the CIR data can be analyzed in both positive and negative situations, so that the liveness detection results in the vehicle cabin can be comprehensively and accurately determined.

[0107] In this way, the present application can perform frequency domain processing and statistical processing based on the CIR data determined by preprocessing, and perform data statistics based on preset scene feature parameters of the presence or absence of living beings in the vehicle cabin, regardless of whether there is a living being in the vehicle cabin, and finally determine the liveness detection result in the form of a liveness detection signal.

[0108] In some embodiments, the liveness detection method further comprises:

[0109] Sends a liveness detection signal to a matching smart device to inform the user of the presence of living things in the vehicle cabin.

[0110] In some embodiments, the liveness detection device further includes a signal transceiver module, which is used to send a liveness detection signal to a matching smart device to inform the user of the presence of a live body in the vehicle cabin.

[0111] In some embodiments, the processor is further configured to send a living body detection signal to a matching smart device to inform the user of the presence of a living body in the vehicle cabin.

[0112] Specifically, on the basis of the above-mentioned implementation mode, when the liveness detection process is completed and the liveness detection signal is generated, in order to accurately and timely feedback the liveness detection result to the car owner or the parking lot manager, so that the car owner or the parking lot manager can quickly locate the vehicle in an emergency, in some examples, when the liveness detection result has been determined and the liveness detection signal has been generated, if the liveness detection result is that there is a live body in the vehicle cabin, the vehicle transmits the liveness detection signal to the above-mentioned smart device through the communication connection relationship between the vehicle and the supporting smart device, so that the smart device can prompt through voice, vibration, screen display, etc. according to the liveness detection signal. The above-mentioned smart device can be a smart phone, a tablet computer, a smart watch, a smart large screen of a parking lot, etc. When the smart device prompts, in addition to prompting that there is a live body in the current vehicle cabin, it can also obtain environmental information around the vehicle, the vehicle's own positioning information, etc. through one or more sensors of the vehicle itself, so that the car owner or parking lot manager can quickly locate the vehicle, thereby realizing rapid rescue.

[0113] In this way, the present application can also send the liveness detection result to the smart device through the communication relationship between the smart devices used in conjunction with the vehicle when the liveness detection result is determined, thereby providing timely and accurate reminders to users regarding the problem of living things left behind in the vehicle cabin, thereby reducing the probability of life risks to living things in the vehicle cabin due to forgetfulness.

[0114] It should be further explained that, in the liveness detection device in the embodiment of the present application, the specific hardware implementation of several modules can be implemented through the following examples:

[0115] First, it should be added that, in addition to the above-mentioned multiple modules, illustratively, the liveness detection device also includes a signal transmission module, which is mainly used to transmit UWB pulse signals into the vehicle cabin, thereby triggering the liveness recognition process.

[0116] Next, the signal receiving module, the signal processing module, the signal detection module and the signal transceiver module can be integrated into a microcontroller, wherein:

[0117] The signal receiving module is generally a UWB radar signal receiving module, which is responsible for receiving and judging the liveness detection start / stop command, and after receiving the liveness detection start command, it starts to enter the working state and then starts to receive the UWB pulse signal. After receiving the liveness detection stop command, the UWB radar signal receiving module stops working and enters the idle state.

[0118] The signal processing module packages, filters, and accumulates the echo data of the received transmitted pulse signal after reflection or scattering by obstacles, removes background noise and other preprocessing to improve the accuracy of the echo data. Finally, it performs Fourier transform on the processed echo data to facilitate the next step of data detection.

[0119] The signal detection module performs statistical analysis on the signal after frequency domain processing, and determines the different statistical characteristics of the scene with living beings and the scene without living beings, and uses the obtained statistical characteristics as the criteria for determining whether there is a living being in the subsequent liveness detection.

[0120] The signal transceiver module is generally manifested as a wireless signal transceiver module that transmits the liveness detection results determined by the algorithm module to other devices with Bluetooth communication function through Bluetooth wireless communication, so that other devices can prompt through voice, vibration, screen display, etc. to remind the owner or parking lot manager of the presence of liveness in the vehicle cabin.

[0121] The vehicle in the embodiment of the present application includes the above-mentioned living body detection device or the above-mentioned electronic device.

[0122] The computer-readable storage medium in the embodiments of the present application stores a computer program, and when the computer program is executed by one or more processors, the above method is implemented.

[0123] In the description of this specification, the descriptions with reference to the terms "certain embodiments", "in an example", "exemplarily", etc., mean that the specific features, structures, materials or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0124] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0125] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for detecting a living body, characterized in that: The method is used for a vehicle, and the method comprises: In response to the ultra-wideband pulse signal generated into the interior of the vehicle cabin, obtaining original channel impulse response data fed back from the interior environment of the vehicle cabin; Performing signal preprocessing on the original channel impulse response data to determine channel impulse data to improve the accuracy of the living body detection signal; A living body detection signal is determined according to the channel pulse data to detect whether a living body exists in the vehicle cabin.

2. The method according to claim 1, characterized in that The performing signal preprocessing on the original channel impulse response data to determine the channel impulse data includes: Pre-analysis processing is performed on the original channel impulse response data to determine first channel impulse response data so as to sort out effective signals in the original channel impulse response data.

3. The method according to claim 2, characterized in that The performing signal preprocessing on the original channel impulse response data to determine the channel impulse data also includes: A packet adaptation process is performed on the first channel impulse response data to determine the channel impulse data so that the channel impulse data can be recognized by the vehicle.

4. The method according to claim 2, characterized in that: The performing signal preprocessing on the original channel impulse response data to determine the channel impulse data also includes: An impulse integration process is performed on the first channel impulse response data to determine second channel impulse data so as to amplify a non-noise portion in the first channel impulse response data.

5. The method according to claim 4, characterized in that The performing signal preprocessing on the original channel impulse response data to determine the channel impulse data also includes: An outlier elimination process is performed on the second channel pulse data to determine the third channel pulse data.

6. The method according to claim 5, characterized in that The performing signal preprocessing on the original channel impulse response data to determine the channel impulse data also includes: The third channel pulse data is filtered to determine the channel pulse data so as to remove burrs in the channel pulse data.

7. The method according to claim 1, characterized in that The step of determining a living body detection signal according to the channel pulse data comprises: Performing Fourier transform according to the channel pulse data to determine a response frequency domain signal; The living body detection signal is statistically determined according to the response frequency domain signal and preset scene feature parameters.

8. The method according to claim 1, characterized in that The method further comprises: The living body detection signal is sent to a matching smart device to inform the user of the presence of a living body in the vehicle cabin.

9. A living body detection device, characterized in that: The device is used for a vehicle, and comprises: A signal receiving module, configured to obtain original channel impulse response data fed back from the internal environment of the vehicle cabin in response to an ultra-wideband pulse signal generated to the interior of the vehicle cabin; A signal processing module, used for performing signal preprocessing on the original channel impulse response data to determine channel impulse data so as to improve the accuracy of the living body detection signal; The signal detection module is used to determine a living body detection signal according to the channel pulse data to detect whether there is a living body in the vehicle cabin.

10. The living body detection device according to claim 9, characterized in that: The device also includes: The signal transceiver module is used to send the living body detection signal to a matching smart device to inform the user of the presence of living bodies in the vehicle cabin.

11. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the method according to any one of claims 1 to 8 is implemented.

12. A vehicle, characterized in that: The vehicle includes the living body detection device according to claim 9 or 10, or includes the electronic device according to claim 11.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by one or more processors, the method according to any one of claims 1 to 8 is implemented.

Images