Living body detection method and device, vehicle and storage medium

Abstract

This disclosure relates to a liveness detection method, apparatus, vehicle, and storage medium. The method includes: in response to detecting a vehicle door changing from an open state to a closed state, controlling an in-vehicle speaker to emit a first sound wave signal; controlling an in-vehicle microphone to acquire a second sound wave signal corresponding to the first sound wave signal; determining channel state information of the second sound wave signal based on the first and second sound wave signals; and determining a liveness detection result based on the channel state information of the second sound wave signal, the liveness detection result indicating the presence of a live person inside the vehicle. Thus, by transmitting and receiving sound wave signals through the in-vehicle speaker and microphone, liveness detection can be achieved without additional detection equipment, utilizing existing in-vehicle equipment. Furthermore, the transmission and reception of sound wave signals from the in-vehicle speaker and microphone are less susceptible to interference, which helps ensure the stability and accuracy of liveness detection.

Description

Technical Field

[0001] This disclosure relates to the field of vehicle technology, and in particular to a liveness detection method, apparatus, vehicle, and storage medium. Background Technology

[0002] In a vehicle's smart cockpit, identifying the presence of occupants is crucial for decision-making. For example, the presence of occupants can influence the activation of auxiliary functions such as power supply status, interior lights, air conditioning, and entertainment systems. Current technologies typically use cameras to detect moving objects inside the vehicle, extracting image changes from the video stream to determine movement. However, camera detection is susceptible to lighting conditions and is prone to being blocked, preventing the acquisition of video streams. Summary of the Invention

[0003] To overcome the problems existing in related technologies, this disclosure provides a liveness detection method, apparatus, vehicle, and storage medium.

[0004] According to a first aspect of the present disclosure, a method for detecting liveness is provided, the method comprising:

[0005] In response to detecting that the car door has changed from an open state to a closed state, the system controls the in-vehicle speakers to emit a first sound wave signal.

[0006] The in-vehicle microphone is controlled to collect a second sound wave signal corresponding to the first sound wave signal;

[0007] Based on the first acoustic signal and the second acoustic signal, determine the channel state information of the second acoustic signal;

[0008] Based on the channel state information of the second acoustic signal, a liveness detection result is determined, which is used to indicate whether a live person exists inside the vehicle.

[0009] Optionally, the second acoustic signal includes multiple frames of data, and the channel state information of the second acoustic signal includes channel state information corresponding to the multiple frames of data.

[0010] Optionally, determining the liveness detection result based on the channel state information of the second acoustic signal includes:

[0011] Based on the channel state information corresponding to the multi-frame data, the target dynamic and static state information is determined. The target dynamic and static state information is used to indicate the motion state of an object moving inside the vehicle or to indicate the static state of an object not moving inside the vehicle.

[0012] If the target's static / dynamic state information corresponds to the motion state, it is determined that a living being exists inside the vehicle.

[0013] If the target's dynamic state information indicates that it is in a static state, it is determined that there is no living person inside the vehicle.

[0014] Optionally, the number of microphones is one or more, and each microphone has a corresponding second acoustic signal and channel state information of the second acoustic signal;

[0015] Determining the target's dynamic / static status information based on the channel state information corresponding to the multi-frame data includes:

[0016] For each microphone, a first motion / static state information of the microphone is determined based on the channel state information of the microphone corresponding to the multi-frame data, wherein the first motion / static state information is the motion state or the static state;

[0017] The target dynamic state information is determined based on the first dynamic state information of each microphone.

[0018] Optionally, determining the first dynamic / static state information of the microphone based on the channel state information of the microphone corresponding to the multi-frame data includes:

[0019] Based on the channel state information of the microphone corresponding to the multi-frame data, determine the discreteness of the channel state information corresponding to each frame of data;

[0020] The first dynamic and static state information is determined based on the dispersion.

[0021] Optionally, the channel state information of a frame of data includes channel state information corresponding to each of the multiple channels;

[0022] The step of determining the discreteness of the channel state information corresponding to each frame of data based on the channel state information of the microphone corresponding to the multi-frame data includes:

[0023] For each target frame data, the differential information between the channel state information of the target frame data and the channel state information of the comparison frame data is determined. The time difference between the comparison frame data and the target frame data is a preset step size. Both the target frame data and the comparison frame data are one frame data in the multi-frame data. The differential information includes the differential value corresponding to each channel.

[0024] For each target frame data, the discreteness of the channel state information corresponding to the target frame data is determined based on the differential value contained in the differential information corresponding to the target frame data.

[0025] Optionally, determining the first dynamic / static state information based on the discreteness includes:

[0026] Based on a preset dispersion threshold, a first number of dispersions greater than or equal to the dispersion threshold is determined, and a second number of dispersions less than the dispersion threshold is determined.

[0027] Based on the comparison result between the first quantity and the second quantity, the first dynamic and static state information is determined.

[0028] Optionally, determining the first dynamic / static state information based on the discreteness includes:

[0029] For each frame of data, the second dynamic and static state information corresponding to the frame of data is determined based on the comparison result between the preset discreteness threshold and the discreteness of the channel state information corresponding to the frame of data. The second dynamic and static state information is the motion state or the static state.

[0030] Based on the second motion and stillness state information of each frame of data, state filtering processing is performed on the second motion and stillness state information to obtain the third motion and stillness state information corresponding to each frame of data, wherein the third motion and stillness state information is the motion state or the stillness state.

[0031] Determine the third number of motion states and the fourth number of stationary states in the third dynamic and static state information;

[0032] The first dynamic / static state information is determined based on the comparison result of the third quantity and the fourth quantity.

[0033] Optionally, if the number of microphones is more than one, determining the target dynamic state information based on the first dynamic state information of each microphone includes:

[0034] Based on the first motion / static state information of each microphone, determine the fifth number of motion states and the sixth number of static states in the first motion / static state information.

[0035] Based on the comparison results of the fifth quantity and the sixth quantity, the target dynamic and static state information is determined.

[0036] Optionally, each frame of data included in the first acoustic signal is generated based on the first baseband signal corresponding to each frame;

[0037] The step of determining the channel state information of the second acoustic signal based on the first acoustic signal and the second acoustic signal includes:

[0038] For each frame of data, the signal corresponding to the frame of data in the second acoustic signal is demodulated to obtain the second baseband signal of the frame of data, and the channel state information corresponding to the frame of data is determined based on the first baseband signal and the second baseband signal.

[0039] Optionally, the method further includes:

[0040] Determine the seat recognition result, which is used to indicate whether there is an object located on a seat in the vehicle;

[0041] Based on the liveness detection results and the seat recognition results, it is determined whether there is a live person inside the vehicle.

[0042] Optionally, at least one seat in the vehicle is equipped with a pressure sensor;

[0043] The determination of the seat recognition result includes:

[0044] Acquire the pressure signal collected by each pressure sensor;

[0045] The seat recognition result is determined based on the pressure signal.

[0046] According to a second aspect of the present disclosure, a liveness detection device is provided.

[0047] The device includes:

[0048] The first control module is configured to control the in-vehicle speaker to emit a first sound wave signal in response to detecting that the door has changed from an open state to a closed state.

[0049] The second control module is configured to control the in-vehicle microphone to collect a second sound wave signal corresponding to the first sound wave signal;

[0050] The first determining module is configured to determine the channel state information of the second acoustic signal based on the first acoustic signal and the second acoustic signal.

[0051] The second determining module is configured to determine the liveness detection result based on the channel state information of the second acoustic signal, the liveness detection result being used to indicate whether a live person exists inside the vehicle.

[0052] According to a third aspect of the present disclosure, a vehicle is provided, comprising:

[0053] processor;

[0054] Memory used to store processor-executable instructions;

[0055] The processor is configured to perform the liveness detection method described in the first aspect of this disclosure.

[0056] Optionally, the vehicle further includes:

[0057] A controller is used to control a target device of the vehicle to a target working state when the processor detects a living person in the vehicle, wherein the target device is a device capable of providing auxiliary functions to the vehicle.

[0058] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, having stored thereon computer program instructions that, when executed by a processor, implement the steps of the liveness detection method provided in the first aspect of the present disclosure.

[0059] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects:

[0060] The above technical solution detects the door closing from opening to closing, controlling the in-vehicle speaker to emit a first sound wave signal and controlling the in-vehicle microphone to collect the signal, obtaining a second sound wave signal corresponding to the first sound wave signal. Then, based on the first and second sound wave signals, the channel state information of the second sound wave signal is determined, and the presence of a living person inside the vehicle is identified based on the channel state information. Therefore, by transmitting and receiving sound wave signals through the in-vehicle speaker and microphone, the detection of a living person inside the vehicle can be achieved without the need for additional detection equipment; liveness detection can be achieved based on existing in-vehicle equipment. Furthermore, the transmission and reception of sound wave signals from the in-vehicle speaker and microphone are not easily affected, which helps ensure the stability and accuracy of liveness detection. In addition, liveness detection using sound waves is only performed when the door closes, meaning that sound wave detection is only initiated when necessary, which helps reduce unnecessary losses.

[0061] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0062] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0063] Figure 1 This is a flowchart illustrating a liveness detection method according to an exemplary embodiment.

[0064] Figure 2 This is an exemplary implementation flowchart of a liveness detection method, illustrating the step of determining a liveness detection result according to an exemplary embodiment.

[0065] Figure 3 This is a block diagram illustrating a liveness detection device according to an exemplary embodiment.

[0066] Figure 4This is a block diagram illustrating a vehicle according to an exemplary embodiment. Detailed Implementation

[0067] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0068] It should be noted that all actions involving the acquisition of signals, information, or data in this disclosure are carried out in compliance with the relevant data protection laws and policies of the country where the location is situated, and with authorization from the owner of the relevant device.

[0069] Figure 1 This is a flowchart illustrating a liveness detection method according to an exemplary embodiment, such as... Figure 1 As shown, the method may include steps 11 to 14.

[0070] In step 11, in response to detecting that the door has changed from an open state to a closed state, the speaker inside the vehicle is controlled to emit a first sound wave signal.

[0071] When the door is detected to change from open to closed, it indicates that someone may be getting on or off the vehicle by opening or closing the door. Based on this, the liveness detection method provided in this disclosure can be used to detect liveness inside the vehicle by using sound waves to determine whether there are still active objects inside the vehicle.

[0072] For example, the open and closed status of a car door can be detected using the car door sensors.

[0073] Optionally, the number of speakers can be one or more, and this disclosure is not limited in this regard. It should be noted that if the number of speakers is more than one, each speaker can emit a first sound wave signal in the manner provided in this disclosure.

[0074] The first sound wave signal that controls the in-vehicle speakers can be designed according to actual needs.

[0075] In step 12, the in-vehicle microphone is controlled to acquire a second sound wave signal corresponding to the first sound wave signal.

[0076] The second acoustic signal can be a signal generated by performing at least one operation on the first acoustic signal, such as reflection, diffraction, or scattering.

[0077] Optionally, the number of microphones can be one or more, and this disclosure does not limit this. It should be noted that if the number of microphones is more than one, each microphone corresponds to a acquired second acoustic signal, and each second acoustic signal can be used for subsequent liveness detection based on the method provided in this disclosure.

[0078] It should be noted that the number of speakers and microphones can be the same or different, and this disclosure does not impose any specific limitations on this.

[0079] In step 13, the channel state information of the second acoustic signal is determined based on the first acoustic signal and the second acoustic signal.

[0080] In step 14, the liveness detection result is determined based on the channel state information of the second acoustic signal.

[0081] Among them, the liveness detection results are used to indicate whether there are any living beings inside the vehicle.

[0082] Channel state information reflects changes in the channel through which acoustic signals are transmitted. In the presence of a living being, the movement of that being causes changes in the channel, which in turn causes changes in the channel state information. Therefore, based on the channel state information, the presence or absence of a living being can be determined.

[0083] Therefore, in this disclosure, the liveness detection result used to indicate whether there is a living body inside the vehicle can be determined based on the channel state information corresponding to the second acoustic signal.

[0084] The above technical solution detects the door closing from opening to closing, controlling the in-vehicle speaker to emit a first sound wave signal and controlling the in-vehicle microphone to collect the signal, obtaining a second sound wave signal corresponding to the first sound wave signal. Then, based on the first and second sound wave signals, the channel state information of the second sound wave signal is determined, and the presence of a living person inside the vehicle is identified based on the channel state information. Therefore, by transmitting and receiving sound wave signals through the in-vehicle speaker and microphone, the detection of a living person inside the vehicle can be achieved without the need for additional detection equipment; liveness detection can be achieved based on existing in-vehicle equipment. Furthermore, the transmission and reception of sound wave signals from the in-vehicle speaker and microphone are not easily affected, which helps ensure the stability and accuracy of liveness detection. In addition, liveness detection using sound waves is only performed when the door closes, meaning that sound wave detection is only initiated when necessary, which helps reduce unnecessary losses.

[0085] To enable those skilled in the art to better understand the liveness detection method provided in this disclosure, the various steps provided in this disclosure will be described in more detail below.

[0086] Considering that the movement of a moving object is a continuous action, in order to improve the accuracy of liveness detection, liveness detection can be performed based on channel state information over a period of time. Therefore, in one possible implementation, both the first and second acoustic signals can be multi-frame data, and correspondingly, the channel state information of the second acoustic signal includes the channel state information corresponding to the multi-frame data.

[0087] If the first sound wave signal is a time-domain signal that includes multiple frames of data, the loudspeaker can emit the time-domain signal sequentially and continuously to emit the first sound wave signal.

[0088] In one possible implementation, the first acoustic signal can be generated in the following manner:

[0089] Generate multiple frames of first baseband signals with autocorrelation higher than a preset correlation threshold;

[0090] The first baseband signal is modulated to a preset ultrasonic frequency band to obtain the first acoustic signal.

[0091] For example, the first baseband signal can be a ZC (Zadoff-Chu) sequence because ZC sequences have good autocorrelation. Accordingly, the first baseband signal is a set of complex signal values. For example, the length of the ZC sequence can be set to 85.

[0092] It should be understood that, in practical applications, the first baseband signal can also be other sequences with characteristics such as good autocorrelation, and this disclosure does not specifically limit them.

[0093] After generating the first baseband signal, multiple frames of the first baseband signal can be modulated to a preset ultrasonic frequency band, that is, the first baseband signal is modulated into an ultrasonic frequency band signal, and thus the ultrasonic frequency band signal is used as the first sound wave signal.

[0094] The preset ultrasonic frequency band can be a frequency band that is difficult for the human ear to hear. The first baseband signal is a low-frequency signal. Therefore, in order to avoid interference to the driver and passengers when the speaker emits the first sound wave signal, the low-frequency first baseband signal can be modulated to the preset ultrasonic frequency band to reduce the impact on the driver and passengers in the vehicle.

[0095] For example, the preset ultrasonic frequency band can be 18kHz-20kHz, within which the human ear can hardly hear any sound.

[0096] According to the above embodiment, if the first baseband signal is a set of complex signal values, then after being modulated to a preset ultrasonic frequency band, it will still be a set of complex signal values. When the loudspeaker emits the first sound wave signal, it only needs to send the real part of the complex signal value. Obviously, the signal received by the microphone in subsequent steps is also the real part of the complex signal. Clearly, the first sound wave signal emitted by the loudspeaker is a time-domain signal.

[0097] In this disclosure, each frame of data in the first acoustic signal is generated based on its corresponding first baseband signal; that is, the first baseband signal is adjusted to generate one frame of data. Multiple frames of data can be generated based on the same first baseband signal, meaning that each frame of data corresponds to the same first baseband signal.

[0098] As described above, each frame of data in the first acoustic signal is generated based on the first baseband signal corresponding to each frame. Accordingly, in one possible implementation, step 13, determining the channel state information of the second acoustic signal based on the first and second acoustic signals, may include the following steps:

[0099] For each frame of data, the signal corresponding to the frame of data in the second acoustic signal is demodulated to obtain the second baseband signal of the frame of data, and the channel state information corresponding to the frame of data is determined based on the first baseband signal and the second baseband signal.

[0100] In other words, for each frame of data in the second acoustic signal, the second baseband signal corresponding to that frame of data is obtained through demodulation processing. Then, the channel state information corresponding to that frame of data can be obtained by using the first baseband signal and the second baseband signal.

[0101] It should be noted that the second acoustic signal received by the microphone contains multiple frames of data. The purpose of this disclosure is to determine the corresponding channel state information by using the data of the relevant frames in the second acoustic signal that are related to the first acoustic signal. Therefore, it is also necessary to identify the multiple frames of data in the second acoustic signal to determine which of them are the data that need to participate in the liveness detection, which is the "second acoustic signal corresponding to the first acoustic signal" described in step 12. For example, the correlation between the first and second acoustic signals can be calculated to determine the multiple frames of data in the second acoustic signal that need to participate in the liveness detection. In this disclosure, the description of the multiple frames of data in the second acoustic signal refers to data that has undergone the above determination process.

[0102] For example, determining the channel state information corresponding to the frame data based on the first baseband signal and the second baseband signal can be achieved by determining the frequency response of the baseband signal. That is, the second baseband signal and the first baseband signal corresponding to the frame data are converted from the time domain to the frequency domain, and the frequency domain signal corresponding to the second baseband signal is divided by the frequency domain signal corresponding to the first baseband signal to obtain the aforementioned frequency response. Thus, the aforementioned frequency response can be used as the channel state information of the frame data.

[0103] As mentioned above, both the first and second acoustic signals are multi-frame data, and each frame of data can have a certain sequence length. Therefore, the channel state information can be represented by a matrix.

[0104] For example, suppose the multi-frame data consists of N frames, and the i-th frame is transmitted at time ti, where 1 ≤ i ≤ N, and each frame includes M sampling points (corresponding to the sequence length mentioned above), with each sampling point being a data point, meaning each frame includes M data points (preferably, M = 85). The channel state information of the second acoustic signal can then be represented by the following channel state information matrix:

[0105]

[0106] Where c1, c2...cM are sampling points, i.e., the data included in each frame. Each element in the matrix represents a channel state information corresponding to each data point. For example, S... t1，c1 S represents the channel state information corresponding to the first data in the first frame of data. tN，cM The channel state information that represents the Mth data point of the Nth frame.

[0107] In one possible implementation, step 14, determining the liveness detection result based on the channel state information of the second acoustic signal, may include steps 21 to 23, such as... Figure 2 As shown.

[0108] In step 21, the target dynamic and static state information is determined based on the channel state information corresponding to the multi-frame data.

[0109] In step 22, if the target's dynamic state information indicates a moving state, it is determined that a living being exists inside the vehicle.

[0110] In step 23, if the target's dynamic state information is static, it is determined that there is no living body inside the vehicle.

[0111] Among them, the target dynamic and static state information is used to indicate the motion state of an object that is moving inside the vehicle or to indicate the static state of an object that is not moving inside the vehicle.

[0112] If there is a moving object inside the vehicle, the object's movement (e.g., being in different positions at different times) will affect the signal transmission channel, thus causing changes in the channel state information. Therefore, the dynamic and static state information of the target can be determined by the channel state information corresponding to multiple frames of data.

[0113] As mentioned above, the number of microphones can be one or more, and each microphone has a corresponding second acoustic signal, and thus has channel state information for the corresponding second acoustic signal. Based on this, in one possible implementation, step 21 may include the following steps:

[0114] For each microphone, the first dynamic / static state information of the microphone is determined based on the channel state information of the microphone corresponding to multiple frames of data.

[0115] The target's motion status information is determined based on the initial motion status information from each microphone.

[0116] Among them, the first dynamic and static state information is either a moving state or a stationary state.

[0117] In other words, the first dynamic / static state information of each microphone can be determined based on the channel state information corresponding to multiple frames of data for each microphone, and then:

[0118] If there is only one microphone, the target's motion status information can be determined directly from the microphone's first motion status information.

[0119] If there is more than one microphone, the target's motion and stillness information can be determined by combining the motion and stillness information from more than one microphone to improve accuracy.

[0120] Optionally, for each microphone, the first dynamic / static status information of that microphone can be determined in the following ways:

[0121] The discreteness of the channel state information corresponding to each frame of data is determined based on the channel state information of the microphone corresponding to multiple frames of data.

[0122] Based on the dispersion, the first dynamic and static state information is determined.

[0123] Taking the channel state information matrix provided in the above embodiment as an example, the dispersion of the channel state information corresponding to each frame of data actually indicates the dispersion of the element corresponding to a column in the matrix.

[0124] Alternatively, the dispersion can be determined using information that characterizes the degree of dispersion of the data, depending on the specific needs. For example, dispersion can be variance. Or, for another example, dispersion can be range.

[0125] In one possible embodiment, determining the dispersion of the channel state information corresponding to each frame of data based on the channel state information of the microphone corresponding to multiple frames of data may include the following steps:

[0126] For each target frame data, determine the difference information between the channel state information of the target frame data and the channel state information of the comparison frame data;

[0127] For each target frame data, the dispersion of the channel state information corresponding to the target frame data is determined based on the differential value contained in the differential information corresponding to the target frame data.

[0128] Both the target frame data and the comparison frame data are single frames from a multi-frame dataset, and the time difference between the comparison frame data and the target frame data is a preset step size. Taking the channel state information matrix mentioned above as an example, the channel state information corresponding to the target frame data is a column of data in this channel state information matrix. Furthermore, assuming the preset step size is 2, if the channel state information corresponding to the target frame data is the first column of data in this matrix [S...]... t1，c1 S t1，c2 S t1，cM ] T Then, the channel state information corresponding to the frame data is the data in the third column of this matrix [S]. t3，c1 S t3，c2 S t3，cM ] T .

[0129] Furthermore, the channel state information of a frame of data includes the channel state information corresponding to each of the multiple channels. In other words, in a column of the aforementioned channel state information matrix, each element corresponds to a channel, meaning that a row in the channel state information matrix represents a channel.

[0130] Accordingly, for the target frame data, the difference information between the channel state information of the target frame data and the channel state information of the comparison frame data is determined by subtracting the columns of the above matrix, and the result includes the difference value corresponding to each channel.

[0131] For example, assuming there are N-1 target frame data, then N-1 differential information will be obtained. For these N-1 frames of data (e.g., the first N-1 columns of the channel state information matrix mentioned above), they can be used as target frame data respectively to obtain their respective differential information.

[0132] Optionally, for each target frame data, determining the difference information between the channel state information of the target frame data and the channel state information of the comparison frame data can be achieved through the following steps:

[0133] The channel state information of the microphone corresponding to multiple frames of data is normalized to obtain normalized channel state information corresponding to multiple frames of data.

[0134] Based on a preset step size, the difference value between each target frame data and the normalized channel state information of the corresponding comparison frame data is determined to obtain the aforementioned difference information.

[0135] Specifically, for each microphone, the channel state information corresponding to multiple frames of data is normalized, which can eliminate the influence of signal strength changes and help ensure that the liveness detection method provided in this disclosure can be applied to various environments.

[0136] For example, the channel state information mentioned above can be amplitude. Correspondingly, normalizing the channel state information of the microphone corresponding to multiple frames of data can be done by performing amplitude normalization processing on the channel state information of the microphone corresponding to multiple frames of data. It is worth noting that the channel state information is a complex number. Performing amplitude normalization processing on a complex number means dividing the complex number corresponding to the channel state information by the amplitude value; the result is still a complex number. The difference between these complex numbers is then calculated, and the corresponding amplitude is obtained from the differenced complex number. Therefore, the difference value between the normalized channel state information mentioned above is the amplitude difference.

[0137] Therefore, based on all the target frame data, multiple differential information can be determined. Based on the multiple differential values ​​contained in each differential information, the dispersion of the channel state information of the target frame data can be further calculated. As mentioned above, the dispersion of the channel state information of the target frame data can be calculated by taking the variance, range, etc., from the differential values ​​in the differential information corresponding to the target frame data.

[0138] Preferably, in order to improve the accuracy of the difference information, after obtaining the difference information, the difference values ​​in the difference information can be further filtered by mean filtering in the form of a sliding window, and the result of mean filtering can be used to update the difference information for subsequent discreteness calculation.

[0139] After determining the discreteness of the channel state information corresponding to each frame of data, the first dynamic and static state information can be determined through this discreteness.

[0140] In one possible implementation, determining the first dynamic and static state information based on the dispersion may include the following steps:

[0141] Based on a preset dispersion threshold, determine a first number of dispersions that are greater than or equal to the dispersion threshold, and determine a second number of dispersions that are less than the dispersion threshold.

[0142] Based on the comparison between the first quantity and the second quantity, the first dynamic and static state information is determined.

[0143] If the dispersion of a frame of data is greater than or equal to the dispersion threshold, the frame can be considered to indicate a motion state; if the dispersion of a frame of data is less than the dispersion threshold, the frame can be considered to indicate a stationary state.

[0144] Therefore, the two are compared based on the first quantity indicating the state of motion and the second quantity indicating the state of rest:

[0145] If the first number is greater than the second number, it can be assumed that more frames of data indicate the motion state, and the first dynamic state information can be determined as the motion state.

[0146] If the first quantity is less than the second quantity, it can be assumed that more frames of data indicate a static state, and the first dynamic / static state information can be determined to be a static state.

[0147] In this way, the first dynamic and static state information can be determined more accurately through voting.

[0148] In another possible implementation, determining the first dynamic and static state information based on the dispersion may include the following steps:

[0149] For each frame of data, the second dynamic and static state information corresponding to the frame of data is determined based on the comparison result between the preset discreteness threshold and the discreteness of the channel state information corresponding to the frame of data.

[0150] Based on the second dynamic and static state information of each frame of data, state filtering processing is performed on the second dynamic and static state information to obtain the third dynamic and static state information corresponding to each frame of data.

[0151] Determine the third quantity of motion states and the fourth quantity of stationary states in the third dynamic and static state information;

[0152] Based on the comparison results of the third and fourth quantities, the first dynamic and static state information is determined.

[0153] The second dynamic / static state information can be either a moving state or a stationary state, and the third dynamic / static state information can also be either a moving state or a stationary state.

[0154] In other words, for each frame of data, the dispersion threshold is compared with the dispersion of the corresponding frame of data. If the dispersion of a frame of data is greater than or equal to the dispersion threshold, the frame can be considered to indicate a motion state; if the dispersion of a frame of data is less than the dispersion threshold, the frame can be considered to indicate a stationary state. Thus, the second motion / stationary state information can be determined.

[0155] After obtaining the second dynamic and static state information, state filtering can be performed on the second dynamic and static state information to obtain the processing result, which is the third dynamic and static state information.

[0156] After obtaining the third dynamic and static state information, the first dynamic and static state information is determined based on the third quantity of motion states and the fourth quantity of static states in the third dynamic and static state information, that is:

[0157] If the third number is greater than the fourth number, it can be assumed that more frames of data indicate the motion state, and the first dynamic and static state information can be determined as the motion state.

[0158] If the third number is less than the fourth number, it can be assumed that more frames of data indicate a static state, and the first dynamic / static state information can be determined to be a static state.

[0159] In this way, by processing the second dynamic and static state information to a certain extent, the accuracy of the first dynamic and static state information can be further improved.

[0160] Optionally, after obtaining the second dynamic / static state information, other optimization methods can be used to further optimize the second dynamic / static state information to obtain optimized dynamic / static state information, which is then used to determine the first dynamic / static state information. For example, the second dynamic / static state information can be optimized using a sliding window voting method. That is, a sliding window of a specified length is set, and by sliding the window, a voting decision is made based on the dynamic / static state information (i.e., motion state or static state) within the sliding window (e.g., the decision is based on the more numerous decisions) to determine the dynamic / static state information of that window. Then, the first dynamic / static state information is determined by combining the motion state and static state results from multiple windows. Other optional optimization methods will not be elaborated upon in this disclosure.

[0161] In one possible implementation, if there is only one microphone, the first dynamic / static state information corresponding to that microphone can be directly determined as the target dynamic / static state information.

[0162] In another possible implementation, if there is more than one microphone, determining the target dynamic state information based on the first dynamic state information of each microphone may include the following steps:

[0163] Based on the first dynamic and static state information of each microphone, determine the fifth number of dynamic states and the sixth number of static states in the first dynamic and static state information.

[0164] Based on the comparison results of the fifth and sixth quantities, the target's dynamic and static status information is determined.

[0165] In other words, if the number of fifth microphones is greater than the number of sixth microphones, it can be assumed that more microphones indicate a motion state, and the target's motion / static status information can be determined as a motion state; if the number of fifth microphones is less than the number of sixth microphones, it can be assumed that more microphones indicate a stationary state, and the target's motion / static status information can be determined as a stationary state.

[0166] Optionally, referring to the above-described approach of determining the first dynamic and static state information based on the second dynamic and static state information, optimization methods such as state filtering can be applied to the first dynamic and static state information to improve the accuracy of the target dynamic and static state information. The relevant content has been described above and will not be repeated here.

[0167] In this way, with multiple microphones installed inside the vehicle, a decision-making mechanism can be set up to combine data from multiple microphones to determine whether a living person is inside the vehicle, thereby improving the accuracy of liveness detection.

[0168] Optionally, in Figure 1 Based on the steps shown, the method provided in this disclosure may further include the following steps:

[0169] Determine the seat recognition result;

[0170] Based on the results of liveness detection and seat recognition, it is determined whether there is a living person inside the vehicle.

[0171] The seat recognition result is used to indicate whether there is an object located on a seat inside the vehicle.

[0172] For example, at least one seat in the vehicle may be equipped with a pressure sensor, and the seat recognition result can be determined accordingly:

[0173] Acquire the pressure signal collected by each pressure sensor;

[0174] The seat recognition result is determined based on the pressure signal.

[0175] In other words, each pressure sensor is used to collect pressure signals, and then, based on the pressure signals (or changes in the pressure signals), it can be determined whether there is an object on the corresponding seat. If no object is detected on any seat, the seat identification result can be determined as no object is located on the seat in the vehicle; if any pressure sensor collects a pressure signal indicating that an object is detected on that seat, the seat identification result can be determined as an object is located on the seat in the vehicle.

[0176] Furthermore, based on the liveness detection results and seat recognition results obtained from sound wave signals, it can be determined whether there is a living person inside the vehicle.

[0177] Optionally, a determination strategy can be set for the liveness detection results and the seat recognition results. For example, confidence levels can be configured for both, or a strategy can be set when the two conflict (e.g., which one should be used when the two are inconsistent) to combine the two to determine whether there is a live person in the vehicle.

[0178] For example, if the liveness detection result indicates the presence of a live person in the vehicle, and the seat recognition result indicates the presence of an object located on a seat in the vehicle, then it can be determined that there is a live person in the vehicle.

[0179] This method, in addition to using sound waves for liveness detection, further combines the results of seat recognition to jointly determine whether there are living people inside the vehicle, which helps improve the accuracy of liveness detection inside the vehicle.

[0180] Based on the liveness detection method provided in this disclosure, the vehicle's auxiliary functions can also be controlled based on the results of liveness detection, thereby improving the experience of drivers and passengers. For example, when a live person is detected in the vehicle, the vehicle's air conditioning can be turned on. It should be noted that the control of vehicle auxiliary functions based on the presence or absence of a live person can be flexibly set according to user needs, and this disclosure does not limit this.

[0181] Figure 3 This is a block diagram illustrating a liveness detection device according to an exemplary embodiment. (Refer to...) Figure 3 The device 30 includes:

[0182] The first control module 31 is configured to control the in-vehicle speaker to emit a first sound wave signal in response to detecting that the door has changed from an open state to a closed state.

[0183] The second control module 32 is configured to control the in-vehicle microphone to collect a second sound wave signal corresponding to the first sound wave signal;

[0184] The first determining module 33 is configured to determine the channel state information of the second acoustic signal based on the first acoustic signal and the second acoustic signal;

[0185] The second determining module 34 is configured to determine the liveness detection result based on the channel state information of the second acoustic signal, the liveness detection result being used to indicate whether there is a live person inside the vehicle.

[0186] Optionally, the second acoustic signal includes multiple frames of data, and the channel state information of the second acoustic signal includes channel state information corresponding to the multiple frames of data.

[0187] Optionally, the second determining module 34 includes:

[0188] The first determining submodule is configured to determine target motion state information based on channel state information corresponding to the multi-frame data, wherein the target motion state information is used to indicate the motion state of an object moving inside the vehicle or to indicate the stationary state of an object not moving inside the vehicle.

[0189] The second determining submodule is configured to determine the presence of a living being in the vehicle if the target's static / dynamic state information indicates that it is in motion.

[0190] The third determination submodule is configured to determine that there is no living body inside the vehicle if the target dynamic state information is the static state.

[0191] Optionally, the number of microphones is one or more, and each microphone has a corresponding second acoustic signal and channel state information of the second acoustic signal;

[0192] The first determining submodule includes:

[0193] The fourth determining submodule is configured to, for each microphone, determine the first motion / static state information of the microphone based on the channel state information of the microphone corresponding to the multi-frame data, wherein the first motion / static state information is the motion state or the static state.

[0194] The fifth determining submodule is configured to determine the target motion state information based on the first motion state information of each microphone.

[0195] Optionally, the fourth determining submodule includes:

[0196] The sixth determining submodule is configured to determine the discreteness of the channel state information corresponding to each frame of data based on the channel state information of the microphone corresponding to the multi-frame data.

[0197] The seventh determining submodule is configured to determine the first dynamic and static state information based on the discreteness.

[0198] Optionally, the channel state information of a frame of data includes channel state information corresponding to each of the multiple channels;

[0199] The sixth determining submodule includes:

[0200] The eighth determining submodule is configured to determine, for each target frame data, the differential information between the channel state information of the target frame data and the channel state information of the comparison frame data, wherein the time difference between the comparison frame data and the target frame data is a preset step size, and both the target frame data and the comparison frame data are one frame data in the multi-frame data, and the differential information includes the differential value corresponding to each channel.

[0201] The ninth determining submodule is configured to, for each target frame data, determine the discreteness of the channel state information corresponding to the target frame data based on the differential value contained in the differential information corresponding to the target frame data.

[0202] Optionally, the seventh determining submodule includes:

[0203] The tenth determining submodule is configured to determine, based on a preset dispersion threshold, a first number of dispersions greater than or equal to the dispersion threshold, and a second number of dispersions less than the dispersion threshold.

[0204] The eleventh determination submodule is configured to determine the first dynamic / static state information based on the comparison result between the first quantity and the second quantity.

[0205] Optionally, the seventh determining submodule includes:

[0206] The twelfth determining submodule is configured to determine the second dynamic and static state information corresponding to each frame of data based on the comparison result between a preset discreteness threshold and the discreteness of the channel state information corresponding to the frame of data. The second dynamic and static state information is the motion state or the static state.

[0207] The filtering submodule is configured to perform state filtering processing on the second motion state information based on the second motion state information of each frame of data to obtain the third motion state information corresponding to each frame of data, wherein the third motion state information is the motion state or the stationary state.

[0208] The thirteenth determining submodule is configured to determine the third number of motion states and the fourth number of stationary states in the third dynamic and static state information;

[0209] The fourteenth determining submodule is configured to determine the first dynamic and static state information based on the comparison result of the third quantity and the fourth quantity.

[0210] Optionally, if the number of microphones is more than one, the fifth determining submodule includes:

[0211] The fifteenth determining submodule is configured to determine, based on the first motion / static state information of each microphone, the fifth number of motion states and the sixth number of static states in the first motion / static state information.

[0212] The sixteenth determination submodule is configured to determine the target dynamic and static state information based on the comparison result of the fifth quantity and the sixth quantity.

[0213] Optionally, each frame of data included in the first acoustic signal is generated based on the first baseband signal corresponding to each frame;

[0214] The first determining module 33 is configured to demodulate the signal in the second acoustic signal corresponding to the frame data for each frame of data to obtain the second baseband signal of the frame data, and determine the channel state information corresponding to the frame data based on the first baseband signal and the second baseband signal.

[0215] Optionally, the device 30 further includes:

[0216] The third determining module is configured to determine the seat recognition result, which is used to indicate whether there is an object located on a seat in the vehicle;

[0217] The fourth determining module is configured to determine whether a living person exists inside the vehicle based on the liveness detection result and the seat recognition result.

[0218] Optionally, at least one seat in the vehicle is equipped with a pressure sensor;

[0219] The third determining module includes:

[0220] The acquisition submodule is configured to acquire the pressure signal collected by each pressure sensor.

[0221] The seventeenth determining submodule is configured to determine the seat recognition result based on the pressure signal.

[0222] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0223] This disclosure also provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the steps of the liveness detection method provided in this disclosure.

[0224] This disclosure also provides a vehicle, including:

[0225] processor;

[0226] Memory used to store processor-executable instructions;

[0227] The processor is configured to execute the liveness detection method provided in any embodiment of this disclosure.

[0228] Alternatively, the vehicle provided in this disclosure may also include:

[0229] The controller is used to control the target equipment of the vehicle to the target working state when the processor detects the presence of a living person inside the vehicle.

[0230] The target device is a device that can provide auxiliary functions for the vehicle.

[0231] For example, vehicle auxiliary functions may include, but are not limited to, the following: whether the power supply is on, whether the interior lights are on, whether the air conditioning is on, whether the entertainment system is on, and whether the anti-theft system is on. Correspondingly, the target devices are: power supply, interior lights, air conditioning system, entertainment system, and anti-theft system, etc. Thus, the target operating states are, in order: power on or off, on or off, on or off, on or off, on or off.

[0232] Each target device can have its own target operating state set according to the actual scenario or requirements. For example, if the interior lights need to be turned on when a living person is in the vehicle, the target operating state of the interior lights can be set to "on". Then, when the processor detects a living person in the vehicle, the controller can control the interior lights to turn on. Similarly, if the air conditioning needs to be turned on when a living person is in the vehicle, the target operating state of the air conditioning system can be set to "on". Then, when the processor detects a living person in the vehicle, the controller can control the air conditioning system to turn on. And if the anti-theft function needs to be disabled when a living person is in the vehicle, the target operating state of the anti-theft system can be set to "disabled". Then, when the processor detects a living person in the vehicle, the controller can control the anti-theft system to disable.

[0233] It should be noted that a vehicle can have various assistance functions, and correspondingly, various target devices. Based on this, the controller can control multiple target devices to be in their respective target operating states. For example, when the processor determines that a living person is present in the vehicle, the controller can activate all of the vehicle's assistance functions.

[0234] This setting allows for the control of vehicle assistance functions based on the results of liveness detection. For example, when a live person is detected in the vehicle, the vehicle's assistance functions can be activated to enhance the experience for drivers and passengers.

[0235] Figure 4 This is a block diagram illustrating a vehicle 600 according to an exemplary embodiment. For example, vehicle 600 can be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicle. Vehicle 600 can be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.

[0236] Reference Figure 4The vehicle 600 may include various subsystems, such as an infotainment system 610, a perception system 620, a decision control system 630, a drive system 640, and a computing platform 650. The vehicle 600 may also include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and each component of the vehicle 600 can be interconnected via wired or wireless means.

[0237] In some embodiments, the infotainment system 610 may include a communication system, an entertainment system, and a navigation system, etc.

[0238] The perception system 620 may include several sensors for sensing information about the environment surrounding the vehicle 600. For example, the perception system 620 may include a global positioning system (which may be GPS, BeiDou, or other positioning systems), an inertial measurement unit (IMU), lidar, millimeter-wave radar, ultrasonic radar, and a camera device.

[0239] The decision control system 630 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0240] The drive system 640 may include components that provide powered motion to the vehicle 600. In one embodiment, the drive system 640 may include an engine, an energy source, a transmission system, and wheels. The engine may be one or a combination of internal combustion engines, electric motors, and compressed air engines. The engine is capable of converting energy provided by the energy source into mechanical energy.

[0241] Some or all of the functions of vehicle 600 are controlled by computing platform 650. Computing platform 650 may include at least one processor 651 and memory 652, processor 651 can execute instructions 653 stored in memory 652.

[0242] Processor 651 can be any conventional processor, such as a commercially available CPU. The processor may also include, for example, a Graphics Processing Unit (GPU), a Field Programmable Gate Array (FPGA), a System on Chip (SOC), an Application Specific Integrated Circuit (ASIC), or a combination thereof.

[0243] The memory 652 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0244] In addition to instruction 653, memory 652 can also store data, such as road maps, route information, vehicle position, direction, speed, and other data. The data stored in memory 652 can be used by computing platform 650.

[0245] In this embodiment of the disclosure, the processor 651 may execute instructions 653 to complete all or part of the steps of the above-described liveness detection method.

[0246] In another exemplary embodiment, a computer program product is also provided, the computer program product comprising a computer program executable by a programmable device, the computer program having a code portion for performing the above-described liveness detection method when executed by the programmable device.

[0247] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of this disclosure. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0248] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for detecting liveness, characterized in that, The method includes: In response to detecting that the car door has changed from an open state to a closed state, the system controls the in-vehicle speakers to emit a first sound wave signal. The in-vehicle microphone is controlled to collect a second sound wave signal corresponding to the first sound wave signal, and the number of microphones is more than one. Based on the first acoustic signal and the second acoustic signal, the channel state information of the second acoustic signal is determined. The second acoustic signal includes multiple frames of data, and the channel state information of the second acoustic signal includes channel state information corresponding to the multiple frames of data. For each microphone, a first motion / static state information of the microphone is determined based on the channel state information of the microphone corresponding to the multi-frame data, wherein the first motion / static state information is a motion state or a static state. Based on the first motion / static state information of each microphone, determine the fifth number of motion states and the sixth number of static states in the first motion / static state information. Based on the comparison results of the fifth quantity and the sixth quantity, target dynamic and static state information is determined. The target dynamic and static state information is used to indicate the motion state of an object that is moving inside the vehicle or to indicate the static state of an object that is not moving inside the vehicle. If the target's static / dynamic state information corresponds to the motion state, it is determined that a living being exists inside the vehicle. If the target's dynamic state information indicates that it is in a static state, it is determined that there is no living person inside the vehicle.

2. The method according to claim 1, characterized in that, Determining the first dynamic / static state information of the microphone based on the channel state information corresponding to the multi-frame data includes: Based on the channel state information of the microphone corresponding to the multi-frame data, determine the discreteness of the channel state information corresponding to each frame of data; The first dynamic and static state information is determined based on the dispersion.

3. The method according to claim 2, characterized in that, The channel state information of a frame of data includes the channel state information corresponding to each of the multiple channels. The step of determining the discreteness of the channel state information corresponding to each frame of data based on the channel state information of the microphone corresponding to the multi-frame data includes: For each target frame data, the differential information between the channel state information of the target frame data and the channel state information of the comparison frame data is determined. The time difference between the comparison frame data and the target frame data is a preset step size. Both the target frame data and the comparison frame data are one frame data in the multi-frame data. The differential information includes the differential value corresponding to each channel. For each target frame data, the discreteness of the channel state information corresponding to the target frame data is determined based on the differential value contained in the differential information corresponding to the target frame data.

4. The method according to claim 2, characterized in that, Determining the first dynamic / static state information based on the discreteness includes: Based on a preset dispersion threshold, a first number of dispersions greater than or equal to the dispersion threshold is determined, and a second number of dispersions less than the dispersion threshold is determined. Based on the comparison result between the first quantity and the second quantity, the first dynamic and static state information is determined.

5. The method according to claim 2, characterized in that, Determining the first dynamic / static state information based on the discreteness includes: For each frame of data, the second dynamic and static state information corresponding to the frame of data is determined based on the comparison result between the preset discreteness threshold and the discreteness of the channel state information corresponding to the frame of data. The second dynamic and static state information is the motion state or the static state. Based on the second motion and stillness state information of each frame of data, state filtering processing is performed on the second motion and stillness state information to obtain the third motion and stillness state information corresponding to each frame of data, wherein the third motion and stillness state information is the motion state or the stillness state. Determine the third number of motion states and the fourth number of stationary states in the third dynamic and static state information; The first dynamic / static state information is determined based on the comparison result of the third quantity and the fourth quantity.

6. The method according to claim 1, characterized in that, The first acoustic signal includes data for each frame generated based on the first baseband signal corresponding to each frame. The step of determining the channel state information of the second acoustic signal based on the first acoustic signal and the second acoustic signal includes: For each frame of data, the signal corresponding to the frame of data in the second acoustic signal is demodulated to obtain the second baseband signal of the frame of data, and the channel state information corresponding to the frame of data is determined based on the first baseband signal and the second baseband signal.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: Determine the seat recognition result, which is used to indicate whether there is an object located on a seat in the vehicle; Based on the liveness detection results and the seat recognition results, it is determined whether there is a live person inside the vehicle.

8. The method according to claim 7, characterized in that, At least one seat in the vehicle is equipped with a pressure sensor; The determination of the seat recognition result includes: Acquire the pressure signal collected by each pressure sensor; The seat recognition result is determined based on the pressure signal.

9. A liveness detection device, characterized in that, The device includes: The first control module is configured to control the in-vehicle speaker to emit a first sound wave signal in response to detecting that the door has changed from an open state to a closed state. The second control module is configured to control the in-vehicle microphones to collect a second sound wave signal corresponding to the first sound wave signal, wherein the number of microphones is more than one. The first determining module is configured to determine the channel state information of the second acoustic signal based on the first acoustic signal and the second acoustic signal, wherein the second acoustic signal includes multiple frames of data, and the channel state information of the second acoustic signal includes channel state information corresponding to the multiple frames of data. The fourth determining submodule is configured to, for each microphone, determine the first dynamic / static state information of the microphone based on the channel state information of the microphone corresponding to the multi-frame data, wherein the first dynamic / static state information is a moving state or a stationary state. The fifteenth determining submodule is configured to determine, based on the first motion / static state information of each microphone, the fifth number of motion states and the sixth number of static states in the first motion / static state information. The sixteenth determining submodule is configured to determine target dynamic and static state information based on the comparison result of the fifth quantity and the sixth quantity. The target dynamic and static state information is used to indicate the motion state of an object moving inside the vehicle or to indicate the static state of an object not moving inside the vehicle. The second determining submodule is configured to determine the presence of a living being in the vehicle if the target's static / dynamic state information indicates that it is in motion. The third determination submodule is configured to determine that there is no living body inside the vehicle if the target dynamic state information is the static state.

10. A vehicle, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to perform the liveness detection method according to any one of claims 1-8.

11. The vehicle according to claim 10, characterized in that, The vehicle also includes: A controller is used to control a target device of the vehicle to a target working state when the processor detects a living person in the vehicle, wherein the target device is a device capable of providing auxiliary functions to the vehicle.

12. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When executed by a processor, the program instructions implement the steps of the method described in any one of claims 1-8.

Images