An in-vehicle life detection method, a detection device, and a computer-readable storage medium

Abstract

This invention relates to a method, device, and computer-readable storage medium for detecting life inside a vehicle. The method includes the following steps: S1, accumulating and acquiring phase matrix signals, processing the complex matrix signals to obtain corresponding signal-to-noise ratio (SNR) distribution data, and determining the presence of a living being based on the SNR distribution data; S2, if a living being is present, performing DBF processing on the complex matrix signals to obtain accumulated moving target angle information at different distance units; S3, performing Cartesian coordinate transformation and generating a three-dimensional spatial point cloud distribution of the SNR based on the SNR distribution data obtained in step S1; S4, statistically analyzing the three-dimensional point cloud distribution to determine the location of the living being inside the vehicle. This invention proposes a method, device, and computer-readable storage medium for detecting life inside a vehicle, which can effectively improve the accuracy of life detection.

Description

Technical Field

[0001] This invention relates to the field of millimeter radar wave application technology, and in particular to a method, detection device, and computer-readable storage medium for in-vehicle life detection. Background Technology

[0002] With the increasing popularity of automobiles, car safety issues have attracted more and more public attention. During vehicle operation, it is necessary to remind passengers in each seat to wear seat belts in order to minimize safety problems caused by vehicle accidents.

[0003] Currently, occupant position detection in vehicles determines the presence of a person in the current seat, thus confirming whether a seatbelt reminder is needed. Existing occupant position detection primarily utilizes infrared sensors, cameras, and pressure sensors. However, infrared sensor and camera technologies are significantly affected by factors such as vehicle interior temperature, obstructions, and light intensity. Pressure sensors cannot distinguish living beings, are easily interfered with by static false targets such as suitcases and backpacks, and have a limited detection range, only able to detect the occupancy of a single corresponding seat. Millimeter-wave radar, due to its high detection accuracy, wide field of view (FOV), and low power consumption, has potential application value in vehicle occupant position detection. Summary of the Invention

[0004] To address the aforementioned problems in existing technologies, this invention proposes a method, detection device, and computer-readable storage medium for in-vehicle life detection. Based on millimeter-wave radar, it can accurately detect whether there are people inside the vehicle and obtain the number and location of the people inside, thus possessing the advantage of high detection accuracy.

[0005] Specifically, this invention proposes a method for detecting life inside a vehicle, which includes a millimeter-wave radar installed inside the vehicle for scanning the interior space, comprising the following steps:

[0006] S1, acquire an intermediate frequency signal and perform FFT processing on the intermediate frequency signal to obtain a set of complex signals; acquire multiple sets of complex matrix signals with different distance units and time units through slow time data accumulation; process the complex matrix signals to obtain the corresponding signal-to-noise ratio distribution data, and determine whether there is a living organism based on the signal-to-noise ratio distribution data;

[0007] S2; If a living organism exists, perform DBF processing on the complex matrix signal obtained in step S1 to obtain the cumulative moving target angle information of different distance units;

[0008] S3, after screening based on the moving target angle information, perform rectangular coordinate transformation, and generate a three-dimensional spatial point cloud distribution of signal-to-noise ratio based on the signal-to-noise ratio distribution data obtained in step S1;

[0009] S4, statistical analysis of the three-dimensional point cloud distribution to determine the location of the living organism inside the vehicle.

[0010] According to one embodiment of the present invention, processing the complex matrix signal to obtain corresponding signal-to-noise ratio (SNR) distribution data, and determining the presence of a living organism based on the SNR distribution data includes the following steps:

[0011] S11, Perform FFT processing on each distance unit of the complex matrix signal to establish a phase change frequency spectrum;

[0012] S12, obtain the power distribution of the target frequency band based on the phase change frequency spectrum, and obtain the signal-to-noise ratio distribution data of each distance unit based on the power distribution of the target frequency band, wherein the target frequency band is used to characterize the breathing frequency band of a life group containing different kinds of life forms.

[0013] S13, the number of effective motion distance units is obtained by screening the signal-to-noise ratio distribution data. The effective motion distance units are used to characterize the possible location of a living organism in a breathing state. The existence of a corresponding living organism is determined based on the number of effective motion distance units.

[0014] According to one embodiment of the present invention, the signal-to-noise ratio distribution data is the average signal-to-noise ratio. In step 12, a slow-time sliding window algorithm is used to obtain multiple sets of signal-to-noise ratios based on the target frequency band, and the average signal-to-noise ratio corresponding to each distance cell is calculated sequentially.

[0015] According to one embodiment of the present invention, the acquisition of each signal-to-noise ratio includes the following steps: taking the ratio of the maximum power value in the target frequency band to the average or median power value of each frequency band as the signal-to-noise ratio of the corresponding range cell.

[0016] According to one embodiment of the present invention, in step S13, the detection result of living organisms is determined by counting the number of distance cells that meet the signal-to-noise ratio threshold.

[0017] According to one embodiment of the present invention, in step S13, at least two sets of signal-to-noise ratio (SNR) ladders are established, and the SNR distribution data is screened through the SNR ladders to obtain the number of effective motion distance units.

[0018] According to one embodiment of the present invention, in step S2, the canceled complex matrix signal with an interval of N frames is used to perform DBF processing to obtain the single moving target angle information of each range unit. After multiple accumulations, the cumulative moving target angle information of different range units is obtained.

[0019] According to one embodiment of the present invention, in step S4, the three-dimensional spatial point cloud distribution is divided using a pre-set rectangular region based on the in-vehicle spatial location, the number of points in each rectangular region is counted, and the count is compared with a threshold value set for the number of points in each rectangular region to determine the rectangular region where the living being is located.

[0020] According to an embodiment of the present invention, the in-vehicle life detection method further includes step S5, which performs occupancy assistance judgment based on signal-to-noise ratio distribution data to determine whether the seat near the millimeter-wave radar side is occupied, and combines the judgment result obtained in step S4 to determine the position of the living body in the vehicle space, and takes the result of occupancy assistance judgment as the standard.

[0021] The present invention also provides an in-vehicle life detection device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any of the aforementioned detection methods.

[0022] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the preceding detection methods.

[0023] This invention provides a method, device, and computer-readable storage medium for in-vehicle life detection. Utilizing millimeter-wave radar technology, it acquires multiple sets of complex matrix signals with different distance and time units through slow-time data accumulation. The complex matrix signals are processed to obtain corresponding signal-to-noise ratio (SNR) distribution data. Based on this SNR distribution data, the presence of a living being is determined. If a living being is found, the acquired complex matrix signals undergo DBF (Deep Frame Function) processing to obtain cumulative moving target angle information at different distance units. After screening, a Cartesian coordinate transformation is performed to generate a three-dimensional spatial point cloud distribution of the SNR, thereby determining the location of the living being inside the vehicle. This detection method effectively obtains the number and location of people inside the vehicle with high accuracy. Furthermore, it improves the accuracy of life detection, solving the problems of false alarms and missed alarms caused by signal interference or reduced signal strength in millimeter-wave radar during in-vehicle life detection.

[0024] It should be understood that the above general description and the following detailed description of the invention are exemplary and illustrative, and are intended to provide further explanation of the invention as described in the claims. Attached Figure Description

[0025] The accompanying drawings are included to provide further explanation of the invention. They are incorporated into and constitute a part of this application. The drawings illustrate embodiments of the invention and, together with this specification, serve to explain the principles of the invention.

[0026] In the attached image:

[0027] Figure 1 A flowchart of an embodiment of the in-vehicle life detection method of the present invention is shown.

[0028] Figure 2A The phase change frequency spectrum is shown in the absence of strong reflection interference and inanimate targets.

[0029] Figure 2B The phase change frequency spectrum is shown when there is no strong reflection interference and a living target is present.

[0030] Figure 2C The frequency spectrum of phase change is shown when there is raincoat interference and no living target.

[0031] Figure 3A This is a diagram showing a person leaning over the left-hand window, using a slow-time sliding window algorithm to obtain multiple signal-to-noise ratios based on the target frequency band.

[0032] Figure 3B Based on Figure 3A A schematic diagram showing the average signal-to-noise ratio for each distance cell.

[0033] Figure 4A This is a schematic diagram showing how a fan is turned on and placed on the back seat, and how a slow-time sliding window algorithm is used to obtain multiple signal-to-noise ratios based on the target frequency band.

[0034] Figure 4B Based on Figure 4A A schematic diagram showing the average signal-to-noise ratio for each distance cell.

[0035] Figure 5A This is a schematic diagram showing how multiple signal-to-noise ratios based on the target frequency band are obtained by using a slow-time sliding window algorithm when an infant is sitting in the left rear child seat.

[0036] Figure 5B Based on Figure 5A A schematic diagram showing the average signal-to-noise ratio for each distance cell.

[0037] Figure 6A A schematic diagram of the point cloud distribution in the 7-coordinate system is shown. Figure 1 .

[0038] Figure 6B The second diagram shows the distribution of point clouds in the coordinate system.

[0039] Figure 6C The diagram shows the point cloud distribution in the coordinate system.

[0040] Figure 6D The diagram below shows the point cloud distribution in the coordinate system.

[0041] Figure 6EThe diagram below shows the point cloud distribution in the coordinate system.

[0042] Figure 6F The diagram shows the point cloud distribution in the coordinate system. (Sixth diagram) Detailed Implementation

[0043] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0044] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this application or its application or use. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0045] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0046] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of this application. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following drawings denote similar items; therefore, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

[0047] In the description of this application, it should be understood that the orientation or positional relationship indicated by directional terms such as "front, back, up, down, left, right", "horizontal, vertical, horizontal" and "top, bottom" is usually based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing this application and simplifying the description. Unless otherwise stated, these directional terms do not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the scope of protection of this application; the directional terms "inner" and "outer" refer to the inner and outer contours relative to the outline of each component itself.

[0048] Furthermore, it should be noted that the use of terms such as "first" and "second" to define components is merely for the purpose of distinguishing the corresponding components. Unless otherwise stated, these terms have no special meaning and therefore should not be construed as limiting the scope of protection of this application. In addition, although the terminology used in this application is selected from commonly known and used terms, some terms mentioned in this application's specification may have been chosen by the applicant according to his or her judgment, and their detailed meanings are explained in the relevant sections of this description. Moreover, this application should be understood not only through the actual terms used, but also through the meaning implied by each term.

[0049] Figure 1 A flowchart of an embodiment of the in-vehicle life detection method of the present invention is shown. As shown, an in-vehicle life detection method includes a millimeter-wave radar for scanning the in-vehicle space, comprising the following steps:

[0050] S1. The transmitted and received waveform signals acquired by the millimeter-wave radar are mixed and filtered to obtain an intermediate frequency (IF) signal. This IF signal is then subjected to an FFT to obtain a set of complex signals. Multiple sets of complex matrix signals with different distance and time units are acquired through slow-time data accumulation. The complex matrix signals are processed to obtain the corresponding signal-to-noise ratio (SNR) distribution data, and the presence of a living organism is determined based on this SNR distribution data. It should be noted that in this embodiment, the millimeter-wave radar is installed next to the left armrest of the rear roof. The processing of the signals acquired by the millimeter-wave radar and subsequent embodiments are described based on the installation location of the millimeter-wave radar. It is easy to understand that this installation location is only illustrative and not limiting. If the millimeter-wave radar is installed next to the right armrest of the rear roof or in other locations, those skilled in the art can use the in-vehicle life detection method provided by this invention to improve the accuracy of life detection.

[0051] S2; If a living organism is present, perform DBF processing on the complex matrix signal obtained in step S1 to obtain the cumulative moving target angle information for different distance units. The accumulated moving target angle information prepares for subsequent moving target angle information screening.

[0052] S3, after screening based on the moving target angle information, a Cartesian coordinate transformation is performed, and a three-dimensional spatial point cloud distribution of the signal-to-noise ratio is generated by combining the signal-to-noise ratio distribution data obtained in step S1. Through screening, the moving target angle information that meets the screening conditions is transformed into a coordinate system, thereby obtaining a three-dimensional spatial point cloud distribution related to living organisms.

[0053] S4, statistical analysis of the three-dimensional point cloud distribution to determine the location of the living organism inside the vehicle.

[0054] Preferably, processing the complex matrix signal to obtain the corresponding signal-to-noise ratio (SNR) distribution data, and determining the presence of a living organism based on the SNR distribution data, includes the following steps:

[0055] S11, Perform FFT processing on each distance cell of the acquired complex matrix signal to establish the phase change frequency spectrum;

[0056] S12: Obtain the power distribution of the target frequency band based on the phase change frequency spectrum, and obtain the signal-to-noise ratio (SNR) distribution data for each range cell based on the power distribution of the target frequency band. The target frequency band is used to characterize the respiratory frequency range of life groups containing different types of organisms. Generally, the respiratory rate of an adult is 12–20 breaths / min, varying with age. Younger children have faster respiratory rates; newborns typically have a respiratory rate of 40–45 breaths / min, sometimes reaching 60 breaths / min. Dogs have a respiratory rate of 20–30 breaths / min, and cats have a respiratory rate of 30–40 breaths / min. Since the breathing of any life group that may be present inside the vehicle will cause changes in their body displacement, and these displacement changes correspond one-to-one with the phase changes of the millimeter-wave radar detection signal, the SNR distribution data for each range cell is obtained through the power distribution of the respiratory frequency band of the life group.

[0057] S13: The number of effective motion distance units is obtained by screening the signal-to-noise ratio distribution data. Effective motion distance units are used to characterize the possible locations of living organisms in a breathing state. The presence of a corresponding living organism is determined based on the number of effective motion distance units. The screening step involves using signal-to-noise ratio distribution data to identify moving targets that match the target life group.

[0058] Preferably, after step S11 and before step S12, fundamental frequency interference in the phase-change frequency spectrum is removed based on the target frequency band. This involves removing interference from highly reflective static targets inside the vehicle to ensure the accuracy of subsequent calculations. Static targets refer to inanimate objects. More preferably, removing fundamental frequency interference includes determining peak descent within the target frequency band. If the peak is in a continuously decreasing state, the maximum power value within that target frequency band is set to zero, and the signal-to-noise ratio of the corresponding range cell is set to zero. Figure 2AThe phase change frequency spectrum is shown in the absence of strong reflection interference and inanimate targets. Figure 2B The phase change frequency spectrum is shown when there is no strong reflection interference and a living target is present. Figure 2C The phase change frequency spectrum is shown when there is raincoat interference and no living target. Figures 2A to 2C In the diagram, the X-axis represents frequency, and the Y-axis represents power. Specifically, the method for removing fundamental frequency interference mainly determines whether a rising segment exists in the first four frequency units, referring to... Figure 2A If there is no rising segment in the first four frequency units and the peaks continuously decline, then the maximum power value of that distance unit is set to 0, and the signal-to-noise ratio of the corresponding distance unit is also set to 0. (Reference) Figure 2B If an increase occurs between the first and second frequency units, the maximum power value and the average power value in the third to sixth frequency units can be used as the basis for subsequent signal-to-noise ratio calculations. The third to sixth frequency units represent the effective frequency range for living targets. Typically, fundamental frequency interference manifests as a high power value at 0 GHz with a wide 0 GHz lobe, affecting the subsequent low-frequency region of the target. (Reference) Figure 2C When interference from the raincoat was present, the power values ​​of the 2nd, 3rd, and 4th frequency units were all increased. Based on the continuous decline of the peak, no significant rising peak was observed within the movement frequency range. Therefore, it was determined that fundamental frequency interference existed in this distance unit, and the maximum power value of this distance unit was set to 0, along with the corresponding signal-to-noise ratio (SNR). As an example, and not a limitation, the presence of fundamental frequency interference can also be determined by using the inflection point or the rate of change of the differential value.

[0059] Preferably, the signal-to-noise ratio (SNR) distribution data is the mean SNR. In step S12, a slow-time sliding window algorithm is used to obtain multiple sets of SNR based on the target frequency band, and the mean SNR corresponding to each range cell is calculated sequentially. More preferably, obtaining each set of SNR includes the following steps: using the ratio of the maximum power value in the target frequency band to the mean power value of each frequency band as the SNR of the corresponding range cell.

[0060] Figure 3A This is a diagram showing a person leaning over the left-hand window, using a slow-time sliding window algorithm to obtain multiple signal-to-noise ratios based on the target frequency band. Figure 3B Based on Figure 3A A schematic diagram showing the average signal-to-noise ratio for each distance cell. Figure 4A This is a schematic diagram showing how a fan is turned on and placed on the back seat, and how a slow-time sliding window algorithm is used to obtain multiple signal-to-noise ratios based on the target frequency band. Figure 4B Based on Figure 4A A schematic diagram showing the average signal-to-noise ratio for each distance cell. Figure 5A This is a schematic diagram showing how multiple signal-to-noise ratios based on the target frequency band are obtained by using a slow-time sliding window algorithm when an infant is sitting in the left rear child seat. Figure 5B Based on Figure 5A A diagram illustrating the average signal-to-noise ratio (SNR) for each distance cell. Figure 3A , 4A In 5A, the X-axis represents the distance cell, the Y-axis represents the slow-time sliding window sequence, and the Z-axis represents the signal-to-noise ratio.

[0061] Due to the poor signal-to-noise ratio stability of interference from both outside and inside the vehicle. Figure 3A The display shows that when there is an interfering target outside the left window, 40 sets of signal-to-noise ratios (SNRs) based on the target frequency band are obtained by performing a slow-time sliding window every 10 frames. In a single SNR test, several values ​​greater than 4 are observed near the 15th distance cell, but the stability is poor. (Reference) Figure 3B After averaging, the average signal-to-noise ratio (SNR) of the 15th range cell is much less than 3, therefore it can be eliminated as an unstable interference signal. Specifically, while targets outside the vehicle can usually be illuminated by millimeter-wave radar, the small size of the windows causes significant obstruction. Even at relatively close distances, the detected SNR data is significantly lower than that of living targets at the same distance inside the cabin. Therefore, living targets outside the window can be eliminated based on the correspondence between different range cells and SNR data. This correspondence between range cells and SNR thresholds can be a dynamic inverse linear relationship or a threshold set empirically in segments. It should be noted that a mean SNR greater than 4 and a range cell greater than 2 are generally considered to indicate the possible presence of a living target; if the mean SNR is not greater than 4, it is considered that no corresponding living target exists.

[0062] Figure 4A The display shows that when handheld fan interference is present in the vehicle, 40 sets of signal-to-noise ratios (SNRs) based on the target frequency band are obtained using a slow-time sliding window algorithm. Several SNR values ​​greater than 4 are observed near the 35th distance cell in a single reading, but the stability is poor. (Reference) Figure 4B After averaging, the mean signal-to-noise ratio (SNR) of the 35th distance cell is much less than 4, indicating that fan interference can be eliminated. The SNR of small or partially obscured in-vehicle life targets remains relatively stable. Figure 5A The simulation of a baby (with respiratory movements) sitting in the child seat on the left side shows a relatively stable signal-to-noise ratio (SNR). Forty SNR values ​​based on the target frequency band were obtained using a slow-time sliding window algorithm. Within a single measurement, multiple SNR values ​​greater than 4 were observed between the 20th and 35th distance units. (Reference) Figure 5B After averaging, it was found that there were multiple signal-to-noise ratios greater than 5 between the 20th and 35th distance units, which retained useful signals as targets for microorganisms.

[0063] As shown above, the slow-time sliding window algorithm can quickly acquire multiple sets of signal-to-noise ratio (SNR) distribution data and sequentially calculate the SNR mean for each distance cell. Since the SNR stability of external and internal vehicle interference is poor, this unstable interference signal is eliminated after averaging. However, the SNR stability of small or partially obscured internal targets is relatively good; taking multiple SNR averages can effectively remove interference signals while retaining the useful signal of small targets. Therefore, by using a sliding window to quickly acquire multiple sets of SNR data, averaging them to filter out interference signals from inside and outside the vehicle, and retaining the useful signal, the accuracy of detecting internal targets can be improved. As an example, and not a limitation, other methods can be used, such as alternating frames or continuous sliding windows, and methods such as weighted summation, median, low-pass filtering followed by averaging or median averaging can be used to obtain the final SNR distribution data.

[0064] Preferably, in step S13, the detection result of a living organism can be determined by counting the number of distance units that meet the signal-to-noise ratio (SNR) threshold. More preferably, in step S13, at least two sets of SNR ladders are established, and the number of valid moving distance units is obtained by screening the SNR distribution data through the SNR ladders. Specifically, the SNR of each set of SNR ladders is snr. N The signal-to-noise ratio is greater than that of SNR. N The distance unit is denoted as the effective motion distance unit of the Nth group of signal-to-noise ratio steps. Let M be the number of effective motion distance units of the Nth group of signal-to-noise ratio steps. N The determination of the existence of a corresponding life form based on the number of effective motion distance units in the signal-to-noise ratio (SNR) step sequence includes: if the number of effective motion distance units meets a preset condition, then the existence of a corresponding life form is determined; the preset condition is: PN≤snrN, QN≤MN; if more than one set of snrN and MN meets the preset condition, then the existence of a corresponding life form is preliminarily determined. Here, PN is the SNR threshold for each SNR step sequence, and QN is the threshold for the number of effective motion distance units in each SNR step sequence. Because the size distribution of life forms inside the vehicle is wide, ranging from large adults to small infants and pets, and infants or pets are often obstructed by seats (including child seats), the SNR distribution range of life forms detected by frequency screening is large, and the number of detected motion distance units also varies greatly. Infants may have fewer effective motion distance units, but the SNR in their corresponding effective motion distance units may be relatively high; adults may have more effective motion distance units, but their corresponding SNR may be lower. To ensure the effectiveness of identifying various living targets, multiple sets of signal-to-noise ratio tiered screening thresholds and effective motion unit number screening thresholds are used to screen living targets.

[0065] In step S13 of an embodiment of the present invention, two groups of signal-to-noise ratio ladders are established, a first signal-to-noise ratio ladder and a second signal-to-noise ratio ladder. The signal-to-noise ratio of the first signal-to-noise ratio ladder is snr1, and the range cells with a signal-to-noise ratio greater than snr1 are recorded as the effective moving range cells of the first signal-to-noise ratio ladder. The signal-to-noise ratio of the second signal-to-noise ratio ladder is snr2, and the range cells with a signal-to-noise ratio greater than snr2 are recorded as the effective moving range cells of the second signal-to-noise ratio ladder. Let the number of effective moving range cells of the first signal-to-noise ratio ladder be M1, and the number of effective moving range cells of the second signal-to-noise ratio ladder be M2. Then, determining whether there is a corresponding living body based on the number of effective moving range cells of the two signal-to-noise ratio ladders includes: if the number of effective moving range cells meets the preset conditions, it is determined that there is a corresponding living body;

[0066] Wherein, the preset conditions are: P1 ≤ snr1, P2 ≤ snr2, P1 < P2, Q1 ≤ M1, Q2 ≤ M2; if snr1 and M1 meet the preset conditions and / or snr2 and M2 meet the preset conditions, it is preliminarily determined that there is a corresponding living body. Let P1 = 4, P2 = 7, then 4 ≤ snr1 < 7, 7 ≤ snr2, Q1 = 5, Q2 = 1. If the number of effective moving range cells M1 of the first signal-to-noise ratio ladder snr1 is 8 and the number of effective moving range cells M2 of the second signal-to-noise ratio ladder snr2 is 1, then Q1(5) ≤ M1(8), Q2(1) ≤ M2(1), and it is preliminarily determined that there is a corresponding living body target for the first signal-to-noise ratio ladder snr1, and there is also a corresponding living body target for the second signal-to-noise ratio ladder snr2; if the number of effective moving range cells M1 of the first signal-to-noise ratio ladder snr1 is 3 and the number of effective moving range cells M2 of the second signal-to-noise ratio ladder snr2 is 3, then Q1(5) ≤ M1(3) is not satisfied, but Q2(1) ≤ M2(3) is satisfied, and it is preliminarily determined that there is no corresponding living body target for the first signal-to-noise ratio ladder snr1, and there is a corresponding living body target for the second signal-to-noise ratio ladder snr2; if the number of effective moving range cells M1 of the first signal-to-noise ratio ladder snr1 is 3 and the number of effective moving range cells M2 of the second signal-to-noise ratio ladder snr2 is 0, then Q1(5) ≤ M1(3) is not satisfied, and Q2(1) ≤ M2(0) is also not satisfied, and it is preliminarily determined that there are no corresponding living body targets for both the first signal-to-noise ratio ladder snr1 and the second signal-to-noise ratio ladder snr2.

[0067] In step S13 of another embodiment of the present invention, three groups of signal-to-noise ratio ladders are established, namely the first signal-to-noise ratio ladder, the second signal-to-noise ratio ladder, and the third signal-to-noise ratio ladder. The signal-to-noise ratio of the first signal-to-noise ratio ladder is snr1, and the range units with a signal-to-noise ratio greater than snr1 are recorded as the effective movement range units of the first signal-to-noise ratio ladder. The signal-to-noise ratio of the second signal-to-noise ratio ladder is snr2, and the range units with a signal-to-noise ratio greater than snr2 are recorded as the effective movement range units of the second signal-to-noise ratio ladder. The signal-to-noise ratio of the third signal-to-noise ratio ladder is snr3, and the range units with a signal-to-noise ratio greater than snr3 are recorded as the effective movement range units of the third signal-to-noise ratio ladder. Let the number of effective movement range units of the first signal-to-noise ratio ladder be M1, let the number of effective movement range units of the second signal-to-noise ratio ladder be M2, and let the number of effective movement range units of the third signal-to-noise ratio ladder be M3. Then, determining whether there is a corresponding living body based on the number of effective movement range units of the three signal-to-noise ratio ladders includes: if the number of effective movement range units meets the preset conditions, it is determined that there is a corresponding living body;

[0068] wherein, the preset conditions are: P1≤snr1, P2≤snr2, P3≤snr3, P1<P2<P3, Q1≤M1, Q2≤M2, Q3≤M3; if snr1 and M1 meet the preset conditions, or snr2 and M2 meet the preset conditions, or snr3 and M3 meet the preset conditions, it is preliminarily determined that there is a corresponding living body. It can be set that P1 = 2, P2 = 3, P3 = 6, Q1 = 6, Q2 = 3, Q3 = 1, corresponding to determining the signal-to-noise ratio ranges of each signal-to-noise ratio ladder, and predicting whether there is a corresponding living body according to the statistical results of the number of effective movement range units corresponding to each signal-to-noise ratio ladder. The specific method refers to the foregoing embodiment and will not be elaborated here.

[0069] Preferably, in step S2, the DBF processing is performed on the cancellation complex matrix signal at an interval of N frames to obtain the single moving target angle information of each range unit, and the cumulative moving target angle information of different range units is obtained after multiple accumulations. Using the cancellation complex matrix signal at an interval of N frames is beneficial for removing the interference of static targets. More preferably, the value of N corresponds to 1 / 4 of the breathing cycle of the living body, so as to obtain more accurate cumulative moving target angle information, and then remove the interference of non-living bodies through subsequent screening of the target angle information.

[0070] Preferably, in step S3, after screening based on the moving target angle information, a Cartesian coordinate transformation is performed, and a three-dimensional spatial point cloud distribution of the signal-to-noise ratio is generated by combining the signal-to-noise ratio distribution data obtained in step S3. The purpose of the screening in this step is to find data that matches the moving target and perform coordinate system transformation to obtain the corresponding three-dimensional spatial point cloud distribution. In the Cartesian coordinate system, the X-axis represents the X direction of the vehicle, and the Y-axis represents the Y direction of the vehicle. Through coordinate system transformation, the point cloud distribution corresponding to the moving target is essentially mapped to the vehicle interior space.

[0071] Preferably, in step S4, the three-dimensional spatial point cloud distribution is divided using pre-set rectangular regions based on the in-vehicle spatial location, the number of points in each rectangular region is counted, and the count is compared with a threshold value set for the number of points in each rectangular region to determine the rectangular region where the living being is located.

[0072] Figure 6A A schematic diagram of the point cloud distribution in the coordinate system is shown. Figure 1 . Figure 6B The second diagram shows the distribution of point clouds in the coordinate system. Figure 6C The diagram shows the point cloud distribution in the coordinate system. Figure 6D The diagram below shows the point cloud distribution in the coordinate system. Figure 6E The diagram below shows the point cloud distribution in the coordinate system. Figure 6F Sixth illustrations show the point cloud distribution in a coordinate system. In the diagram, the X-axis represents the lateral distance inside the vehicle, the Y-axis represents the longitudinal distance, and the Z-axis represents the signal-to-noise ratio (equivalent to the vertical direction inside the vehicle). The 3D spatial point cloud distribution is divided using pre-defined rectangular regions based on the interior spatial locations. The rear passenger space, corresponding to the left, center, and right seats, is divided into three spaces: left, center, and right. A point count threshold is set for each space. The threshold can be set to a value between 5 and 10. (Reference) Figures 6A to 6F Judgment based on a set threshold Figure 6A There are no living organisms inside the car. Figure 6B There is a living organism on the left side. Figure 6C There is a living organism on the right side. Figure 6D There are living organisms in the middle. Figure 6E There are living organisms on the left and right sides. Figure 6F There are living beings on the left, right, and middle sides of the vehicle. If we use a three-digit number to represent the living beings inside the vehicle, with 1 indicating presence and 0 indicating absence, then... Figures 6A to 6F The cases where life forms exist in the left, middle, and right spaces are represented as 000, 100, 001, 010, 101, and 111, respectively.

[0073] Preferably, the in-vehicle life detection method further includes step S5, which performs a occupancy-assisted judgment based on signal-to-noise ratio distribution data to determine whether the seat closest to the millimeter-wave radar is occupied. This is combined with the judgment result obtained in step S4 to determine the location of a living being in the vehicle space, with the occupancy-assisted judgment result being the final determination. If the rear seat space corresponds to the left, middle, and right seats, it is divided into three spaces: left, middle, and right. A three-digit or letter representation is used to indicate the presence of a living being in each space: 1 indicates presence, 0 indicates absence, and X indicates presence or absence. The occupancy-assisted judgment result may be 1XX, 01X, 00X, or 000. If it is 1XX, it indicates that the left-side position closest to the millimeter-wave radar is occupied, meaning there is a living being in the left position, but it cannot be determined whether there are living beings in the middle or right position. If it is 01X, it indicates that the middle position closest to the millimeter-wave radar is occupied, determining that there is no living being in the left position, but it cannot be determined whether there is a living being in the right position. If the value is 00X, it indicates that the right side closest to the millimeter-wave radar is occupied, confirming that there are no living beings in the left and middle positions. If the value is 000, it indicates that there are no living beings in the left, middle, and right positions. When the occupancy assistance result is combined with the result obtained in step S7, the result of the occupancy assistance decision takes precedence. That is, if the occupancy assistance determines whether there is a living being in a certain position, the result of the occupancy assistance decision takes precedence. If the occupancy assistance result is uncertain, the result obtained in step S7 is selected. The table below shows the occupancy status of the vehicle interior space after the combination of the two results.

[0074]

[0075]

[0076] The present invention also provides an in-vehicle life detection device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of any of the aforementioned detection methods.

[0077] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the aforementioned detection methods.

[0078] The specific implementation methods and technical effects of the in-vehicle life detection device and the computer-readable storage medium can be found in the embodiments of the detection method provided by the present invention, and will not be repeated here.

[0079] Those skilled in the art will further appreciate that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps are described above in a generalized manner in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art may implement the described functionality in different ways for each specific application, but such implementation decisions should not be construed as departing from the scope of the invention.

[0080] The various illustrative logic modules and circuits described in conjunction with the embodiments disclosed herein may be implemented or performed using a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. The general-purpose processor may be a microprocessor, but in alternatives, it may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors cooperating with a DSP core, or any other such configuration.

[0081] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of both. The software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor such that the processor can read and write information to / from the storage medium. In an alternative, the storage medium may be integrated into the processor. The processor and storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In an alternative, the processor and storage medium may reside as discrete components in the user terminal.

[0082] In one or more exemplary embodiments, the described functionality may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functionality may be stored or transmitted as one or more instructions or code on or through a computer-readable medium. A computer-readable medium includes both computer storage media and communication media, encompassing any medium that facilitates the transfer of a computer program from one location to another. A storage medium may be any available medium accessible to a computer. By way of example and not limitation, such a computer-readable medium may include RAM, ROM, EEPROM, CD-ROM or other optical disc storage, disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and is accessible to a computer. Any connection is also legitimately referred to as a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of a medium. As used in this article, disk and disc include compact discs (CDs), laser discs, optical discs, digital multi-purpose discs (DVDs), floppy disks, and Blu-ray discs. Disks typically reproduce data magnetically, while discs reproduce data optically using lasers. Combinations of these should also be included within the scope of computer-readable media.

[0083] It will be apparent to those skilled in the art that various modifications and variations can be made to the exemplary embodiments described above without departing from the spirit and scope of the invention. Therefore, it is intended that this invention cover modifications and variations falling within the scope of the appended claims and their equivalents.

Claims

1. A method for detecting life inside a vehicle, comprising a millimeter-wave radar for scanning the interior space of the vehicle, including the following steps: S1, acquire an intermediate frequency (IF) signal and perform FFT processing on the IF signal to obtain a set of complex signals; acquire multiple sets of complex matrix signals with different distance units and time units through slow-time data accumulation; process the complex matrix signals to obtain the corresponding signal-to-noise ratio (SNR) distribution data, and determine the presence of a living organism based on the SNR distribution data, including the following steps: S11, Perform FFT processing on each distance unit of the complex matrix signal to establish a phase change frequency spectrum; S12, obtain the power distribution of the target frequency band based on the phase change frequency spectrum, and obtain the signal-to-noise ratio distribution data of each range cell based on the power distribution of the target frequency band, wherein, The target frequency band is used to characterize the respiratory frequency band of a life group containing different kinds of life forms; S13, the number of effective motion distance units is obtained by screening the signal-to-noise ratio distribution data. The effective motion distance units are used to characterize the possible locations of living beings in a breathing state. The existence of a corresponding living being is determined based on the number of effective motion distance units. S2; If a living organism exists, perform DBF processing on the complex matrix signal obtained in step S1 to obtain the cumulative moving target angle information of different distance units; S3, after screening based on the moving target angle information, perform rectangular coordinate transformation, and generate a three-dimensional spatial point cloud distribution of signal-to-noise ratio based on the signal-to-noise ratio distribution data obtained in step S1; S4, statistical analysis of the three-dimensional point cloud distribution to determine the location of the living organism inside the vehicle.

2. The in-vehicle life detection method as described in claim 1, characterized in that, The signal-to-noise ratio distribution data is the average signal-to-noise ratio. In step 12, a slow-time sliding window algorithm is used to obtain multiple sets of signal-to-noise ratios based on the target frequency band, and the average signal-to-noise ratio corresponding to each distance cell is calculated in sequence.

3. The in-vehicle life detection method as described in claim 2, characterized in that, The acquisition of each signal-to-noise ratio includes the following steps: the ratio of the maximum power value in the target frequency band to the average or median power value of each frequency band is taken as the signal-to-noise ratio of the corresponding distance cell.

4. The in-vehicle life detection method as described in claim 1, characterized in that, In step S13, the number of distance cells that meet the signal-to-noise ratio threshold is counted to determine the life detection result.

5. The in-vehicle life detection method as described in claim 1, characterized in that, In step S13, at least two sets of signal-to-noise ratio (SNR) ladders are established, and the SNR distribution data is screened through the SNR ladders to obtain the number of effective motion distance units.

6. The in-vehicle life detection method as described in claim 1, characterized in that, In step S2, DBF processing is performed on the canceled complex matrix signal at intervals of N frames to obtain the single moving target angle information of each range unit. After multiple accumulations, the cumulative moving target angle information of different range units is obtained.

7. The in-vehicle life detection method as described in claim 1, characterized in that, In step S4, the three-dimensional spatial point cloud distribution is divided using pre-set rectangular regions based on the in-vehicle spatial location. The number of points in each rectangular region is counted and compared with a threshold value set for the number of points in each rectangular region to determine the rectangular region where the living organism is located.

8. The in-vehicle life detection method as described in claim 1, characterized in that, It also includes step S5, which performs a occupancy assistance judgment based on signal-to-noise ratio distribution data to determine whether the seat near the millimeter-wave radar side is occupied. The judgment result obtained in step S4 is combined to determine the position of the living being in the vehicle space, and the result of the occupancy assistance judgment is used as the standard.

9. An in-vehicle life detection device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the in-vehicle life detection method as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the in-vehicle life detection method as described in any one of claims 1-8.

Images