In-vehicle target detection method and device, storage medium and electronic equipment

Abstract

The embodiment of the invention provides an in-vehicle target detection method and device, a storage medium and electronic equipment, and the method comprises the steps: continuously observing the interior of a current vehicle through a vehicle-mounted radar when the current vehicle is in a locked state, and enabling the observation range of the vehicle-mounted radar to cover the seat area of the current vehicle; under the condition that it is determined that an observation target exists in the vehicle according to the observation data in the vehicle, respiratory frequency detection is conducted on the observation target based on the target position of the observation target, and the respiratory frequency of the observation target is obtained; and when the breathing frequency of the observation target is greater than or equal to the specified breathing frequency, carrying out a leaving alarm on the observation target, so as to solve the technical problem of poor target detection accuracy of an in-vehicle target detection method in the related technology, and improve the target detection accuracy.

Description

Technical Field

[0001] Embodiments of the present application relate to the field of vehicle technology, and more specifically, to a method and device for detecting an in-vehicle target, a storage medium, and an electronic device. Background Art

[0002] Vehicles are commonly used means of transportation. In order to ensure the safety of passengers in the vehicle, in-vehicle safety inspections are required. In-vehicle safety inspections can detect children left behind in the vehicle, people with limited behavioral abilities, or other people who are unable to open the vehicle and leave on their own, so as to effectively ensure the safety of passengers in the vehicle.

[0003] At present, the in-vehicle target detection method in the related art is usually implemented by using a contact sensor. For example, a target curve is generated based on the force of the observed target on the seat, and liveness detection is performed based on the target curve. However, the contact detection method is prone to misdetection of targets with light weight, and it is easy to miss detection when the target with light weight is located at the foot pad in the middle of the seat. In addition, heavy objects on the seat will also cause false alarms.

[0004] It can be seen that the in-vehicle target detection method in the related art has the problem of poor target detection accuracy. Summary of the invention

[0005] The embodiments of the present application provide a method and device for detecting an in-vehicle target, a storage medium, and an electronic device to at least solve the technical problem of poor target detection accuracy in the in-vehicle target detection method in the related art.

[0006] According to one aspect of an embodiment of the present application, a method for detecting a target within a vehicle is provided, comprising: when a current vehicle is in a locked state, continuously observing the interior of the current vehicle by means of a vehicle-mounted radar, wherein the observation range of the vehicle-mounted radar covers the seat area of ​​the current vehicle; when it is determined that an observation target exists within the vehicle based on observation data within the vehicle, performing respiratory rate detection on the observation target based on a target position of the observation target to obtain the respiratory rate of the observation target; when the respiratory rate of the observation target is greater than or equal to a specified respiratory rate, issuing a legacy alarm for the observation target.

[0007] According to another aspect of an embodiment of the present application, there is also provided an in-vehicle target detection device, including: an observation unit, for continuously observing the interior of a current vehicle through a vehicle-mounted radar when the current vehicle is in a locked state, wherein the observation range of the vehicle-mounted radar covers the seat area of ​​the current vehicle; a detection unit, for, when it is determined that there is an observation target inside the vehicle according to the observation data inside the vehicle, performing respiratory frequency detection on the observation target based on the target position of the observation target to obtain the respiratory frequency of the observation target; and an alarm unit, for, when the respiratory frequency of the observation target is greater than or equal to a specified respiratory frequency, issuing a legacy alarm to the observation target.

[0008] In an exemplary embodiment, the detection unit includes: an acquisition module, used to acquire the phase time series of the observation target based on the target position of the observation target, wherein the phase time series includes the phase values ​​of the observation target observed at N observation moments within a specified period time; a first execution module, used to perform power spectrum estimation on the phase time series to obtain a target power spectrum, and determine the maximum frequency of the target power spectrum as the breathing frequency of the observation target.

[0009] In an exemplary embodiment, the acquisition module includes: a first extraction submodule, used to extract the target data of the observation target from the observation data of the interior of the vehicle observed by the vehicle-mounted radar at the N observation times based on the target position of the observation target, so as to obtain the target data of the observation target at the N observation times; and a second extraction submodule, used to take the phase of the target data of the observation target at each of the N observation times to obtain the phase time series.

[0010] In an exemplary embodiment, the first execution module includes: a division submodule, used to divide the phase time series into L segments of data, wherein each segment of the L segments of data includes M data in the phase time series, L and M are both positive integers greater than or equal to 2, and two adjacent segments of the L segments of data partially overlap; a conversion submodule, used to use a specified window function to convert each segment of data into a corresponding periodogram, so as to obtain a periodogram corresponding to each segment of data; and a processing submodule, used to average the power spectral density of the same frequency point in the periodogram corresponding to each segment of data, so as to obtain the target power spectrum.

[0011] In an exemplary embodiment, the in-vehicle data of the current vehicle acquired through an observation by the on-board radar is a frame of target observation matrix, and each frame of the target observation matrix observed by the on-board radar is used to record the distance-Doppler matrix observed at the corresponding observation time; the device also includes: an execution unit, which is used to perform peak search on multiple frames of the target observation matrix observed by the on-board radar according to a specified distance interval and a specified speed interval, and determine the position corresponding to the searched peak as the position of one of the observed targets.

[0012] In an exemplary embodiment, the vehicle-mounted radar is a vehicle-mounted millimeter-wave radar; the observation unit includes: a second execution module, used to observe the interior of the vehicle through the vehicle-mounted radar within a radar data frame period of the vehicle-mounted radar, and obtain P receiving channel data of the vehicle-mounted radar, wherein P is a positive integer greater than or equal to 2, and the P receiving channel data include environmental detection information; a transformation module, used to perform a two-dimensional fast Fourier transform on each of the P receiving channel data to obtain P distance-speed information matrices; an accumulation module, used to perform channel accumulation on the P distance-speed information matrices to obtain a frame of the target observation matrix, wherein the channel accumulation refers to the accumulation of elements at the same position in the P distance-speed information matrices.

[0013] In an exemplary embodiment, the alarm unit includes: an alarm module, which is used to issue a legacy alarm in at least one of the following ways when the breathing frequency of the observed target is greater than or equal to the specified breathing frequency and when the distance between the target object bound to the current vehicle and the current vehicle is within a specified distance range: controlling an indicator light arranged on the outside of the current vehicle to flash at a specified frequency to issue a legacy visual alarm, wherein when the indicator light flashes, the color of the indicator light is a specified color, or a plurality of rotating colors; controlling a display screen arranged on the outside of the current vehicle to display a specified graphic or a specified symbol to issue a legacy visual alarm; playing an alarm sound through a sound playing component on the current vehicle to issue a legacy auditory alarm; a sending module, which is used to send an alarm message to a terminal device of the target object to remotely alarm the observed target when the target object bound to the current vehicle is not detected within the specified distance range of the current vehicle.

[0014] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0015] According to another aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the above method embodiments when executing the computer program.

[0016] According to another aspect of the embodiments of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in any of the above method embodiments.

[0017] In an embodiment of the present application, a method of using a vehicle-mounted radar to detect targets and analyzing the breathing frequency of targets in the vehicle to determine targets is adopted. When the current vehicle is in a locked state, the vehicle-mounted radar is used to continuously observe the interior of the current vehicle, wherein the observation range of the vehicle-mounted radar covers the seat area of ​​the current vehicle; when it is determined that there is an observed target inside the vehicle based on the observation data inside the vehicle, the breathing frequency of the observed target is detected based on the target position of the observed target to obtain the breathing frequency of the observed target; when the breathing frequency of the observed target is greater than or equal to the specified breathing frequency, a legacy alarm is issued for the observed target. Since the targets in the vehicle are detected by the vehicle-mounted radar, target detection can be performed by a non-contact detection method, avoiding the reliance on target contact due to the contact detection method, thereby improving the accuracy of target detection; the breathing frequency can effectively distinguish between children, the elderly and other people with limited behavioral abilities, and can effectively identify vital signs signals left in the vehicle, thereby achieving the technical effect of improving the accuracy of in-vehicle target detection, thereby solving the technical problem of poor target detection accuracy in the in-vehicle target detection method in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a schematic diagram of an application scenario of an in-vehicle target detection method according to an embodiment of the present application;

[0019] Figure 2 is a flow chart of an optional in-vehicle target detection method according to an embodiment of the present application;

[0020] Figure 3 is a logic block diagram of an optional child left-behind detection according to an embodiment of the present application;

[0021] Figure 4 is a flow chart of another optional in-vehicle target detection method according to an embodiment of the present application;

[0022] Figure 5 is a flow chart of another optional in-vehicle target detection method according to an embodiment of the present application;

[0023] Figure 6 is a structural block diagram of an optional in-vehicle object detection device according to an embodiment of the present application;

[0024] Figure 7 It is a block diagram of a computer system structure of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] According to one aspect of an embodiment of the present application, a method for detecting an in-vehicle target is provided. Optionally, in this embodiment, the in-vehicle target detection method can be applied to Figure 1 The hardware environment shown includes a vehicle-mounted radar 102, a vehicle-mounted terminal 104 and a server 106. Figure 1 As shown, the vehicle-mounted radar 102 and the vehicle-mounted terminal 104 are located on the same vehicle, and the vehicle-mounted radar 102 can continuously observe the interior of the vehicle and transmit the observation information to the vehicle-mounted terminal 104. The server 106 is connected to the vehicle-mounted terminal 104 through a network and can be used to provide services for the vehicle-mounted terminal 102 (for example, positioning, communication services, etc.).

[0028] In this embodiment, the network may include but is not limited to at least one of the following: wired network, wireless network. The wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, local area network, and the wireless network may include but is not limited to at least one of the following: WIFI (Wireless Fidelity), Bluetooth. The vehicle-mounted radar may include but is not limited to at least one of the following: laser radar, millimeter wave radar, ultrasonic radar.

[0029] The vehicle target detection method in this embodiment can be executed by the vehicle terminal 104 alone, or can be executed by the terminal device 104 in cooperation with at least one of the vehicle radar 102 and the server 106, wherein the vehicle terminal 104 executes the vehicle target detection method in this embodiment by the application running thereon. In the case where the vehicle terminal 104 executes it alone, the vehicle terminal 104 can determine whether there is a target in the vehicle based on the observation information of the vehicle radar 102. In the case where the vehicle terminal 104 and the server 106 execute it together, the vehicle terminal 104 can also send the observation information of the vehicle radar 102 to the server 106, and then the server 106 determines whether there is a target in the vehicle, or after determining that there is a target in the vehicle, the server 106 calls or sends information to the mobile device of the relevant personnel to warn.

[0030] Taking the vehicle terminal 104 executing the vehicle target detection method in this embodiment as an example, Figure 2 is a flow chart of an optional in-vehicle target detection method according to an embodiment of the present application, such as Figure 2 As shown, the process of the above method may include the following steps:

[0031] Step S202, when the current vehicle is in a locked state, continuously observing the interior of the current vehicle through a vehicle-mounted radar, wherein the observation range of the vehicle-mounted radar covers the seat area of ​​the current vehicle;

[0032] Step S204, when it is determined that there is an observation target inside the vehicle according to the observation data inside the vehicle, based on the target position of the observation target, a breathing frequency detection is performed on the observation target to obtain the breathing frequency of the observation target;

[0033] Step S206: When the respiratory frequency of the observed target is greater than or equal to the specified respiratory frequency, a legacy alarm is issued to the observed target.

[0034] The in-vehicle target detection method in this embodiment can be applied to the field of vehicle technology, and can be applied to the scenario of using a vehicle-mounted radar to detect whether there is a designated target in the vehicle. The above-mentioned vehicle-mounted radar can be a millimeter-wave radar, or it can be other laser radars that can observe targets in the vehicle. The above-mentioned designated target can be a child or other living target that is still in the vehicle when the vehicle is currently locked. Living targets left in a locked vehicle may cause dehydration, hypoxia and other problems, thereby endangering the life safety of the above-mentioned living targets and causing serious consequences. Therefore, in order to ensure the safety of the designated target, in-vehicle safety detection and timely alarm after the designated target is detected are particularly important.

[0035] In related technologies, in-vehicle target detection is mostly achieved using contact sensors, vision, and multi-sensor collaborative solutions, which are highly complex and pose a risk of privacy leakage. For example, a target curve is generated based on the force exerted by a specified target on the seat, and liveness detection is performed based on the target curve. However, pressure detection is a contact detection method, which is prone to misdetection of targets with smaller weights, such as children, and is also prone to missed detection when the child is on the foot pad in the middle of the seat.

[0036] To this end, non-contact target detection can be used to overcome the problems of contact target detection. For example, live detection can be achieved by including at least two ultra-wideband radars that support two-way detection to avoid blind spots. However, while the dual-radar method improves accuracy, the hardware cost and complexity of the system are higher, and software processing must also consider data synchronization. The difficulty of troubleshooting is greater than that of a single radar system, and its practicality is affected to a certain extent.

[0037] In this embodiment, the vehicle-mounted radar is used for target detection, and the target is determined by analyzing the breathing frequency of the target in the vehicle. Since the vehicle-mounted radar is used to detect the target in the vehicle, the target detection can be performed through a non-contact detection method, avoiding the reliance on target contact due to the contact detection method. At the same time, compared with other complex detection systems, the hardware cost and complexity of the system can be reduced; and the breathing frequency can effectively distinguish children, the elderly and other people with limited behavioral abilities, and can effectively identify the vital signs left in the car, thereby improving the accuracy of target detection. In addition, the detection system in this embodiment can be compatible with the vehicle-mounted electronic system in the relevant technology, and can be easily integrated into various models, providing a flexible solution for vehicle manufacturing.

[0038] For the current vehicle, in order to determine whether there are designated targets left behind inside the current vehicle (for example, children left behind), when the current vehicle is in a locked state, the interior of the current vehicle can be continuously observed through the on-board radar. Among them, the above-mentioned on-board radar can be at least one radar whose observation range can cover all seating areas of the current vehicle. The on-board radar can be any laser radar capable of target detection, which can be installed at key positions inside the vehicle to cover all possible seating areas. Optionally, the on-board radar can be an on-board millimeter-wave radar, where the on-board millimeter-wave radar has the characteristics of small size, low cost, adaptability to different weather, and can work all day long. As a non-contact detection tool, millimeter-wave radar can be applied to the process of in-vehicle target detection.

[0039] Optionally, the vehicle-mounted radar can adopt the TD-MIMO (Time Division Multiplexing MultipleInput Multiple Output) system to improve the efficiency of target observation. The vehicle-mounted radar using the TD-MIMO system is a radar system that combines time division multiplexing technology and the concept of multiple-input multiple-output radar. It has the advantages of high resolution, high signal-to-noise ratio, strong penetration, good concealment, etc. It can effectively resist interference from complex environments and improve target detection accuracy. In addition, it can continuously send and receive signals to achieve real-time monitoring.

[0040] When the vehicle-mounted radar continuously observes the interior of the current vehicle, observation data of the interior of the vehicle will be generated, and observation data can be obtained each time an observation is made. Since observing the interior of the vehicle is a continuous process, observation data can be continuously obtained. The observation data may include information such as target distance, target speed, and echo intensity. Based on the above observation data, it can be determined whether there is an observation target inside the vehicle, and the target position of the observation target can be determined. Here, the observation data can reflect the status of different positions in the detection space, such as empty, with targets, etc., and different targets are represented differently. Therefore, it can be determined whether there is an observation target inside the vehicle and the location of the observation target, that is, the target position.

[0041] After determining the target position of the observation target, the breathing frequency of the observation target can be detected to obtain the breathing frequency of the observation target. The breathing frequency of the observation target can be detected by frequency domain analysis of the observation data of the vehicle-mounted radar, or by using other detection equipment to detect the breathing frequency of the observation target at the target position, or by other methods, which are not limited in this embodiment. Here, compared with laser radar, the type of information required to be detected is less, and lower-cost detection equipment can be selected to reduce the cost of target detection.

[0042] After obtaining the respiratory frequency of the observed target, the respiratory frequency of the observed target can be compared with the specified respiratory frequency. The specified respiratory frequency is a decision threshold of the set respiratory signal, which can be the respiratory frequency threshold of the living target, the respiratory frequency threshold corresponding to the specified target, the respiratory frequency threshold corresponding to the abnormal state of the specified target, or other respiratory frequency thresholds. In the case where the respiratory frequency of the observed target is greater than or equal to the specified respiratory frequency, it can be confirmed that the observed target is a specified target (for example, a child) left in the car. At this time, a legacy alarm can be issued to the observed target. The legacy alarm is used to prompt the existence of a legacy target in the current vehicle (the alarm is only used for prompting, not for a specific target), and can also be used to prompt that the observed target is left in the current vehicle (the alarm is not only used for prompting, but also for prompting a specific target). There are one or more ways of legacy alarm, which may include but are not limited to at least one of the following: the vehicle indicator light flashes, the vehicle plays an alarm audio, and sends a letter or calls to the driver's mobile device (such as a mobile phone). In addition, there may be other alarm methods, which are not limited in this embodiment.

[0043] For example, if the target is children, Figure 3 As shown, the child left behind detection method mainly includes: data collection; target position estimation; target breathing rate detection; child left behind detection. In addition, when a child is detected to be left behind in the car, a left behind alarm can be executed.

[0044] Through the embodiments provided by the present application, when the current vehicle is in a locked state, the interior of the current vehicle is continuously observed through the on-board radar, wherein the observation range of the on-board radar covers the seat area of ​​the current vehicle; when it is determined that there is an observation target inside the vehicle based on the observation data inside the vehicle, the breathing rate of the observation target is detected based on the target position of the observation target to obtain the breathing rate of the observation target; when the breathing rate of the observation target is greater than or equal to the specified breathing rate, a legacy alarm is issued for the observation target, thereby solving the technical problem of poor target detection accuracy in the in-vehicle target detection method in the related art and improving the accuracy of in-vehicle target detection.

[0045] In an exemplary embodiment, based on the target position of the observation target, the respiratory frequency of the observation target is detected to obtain the respiratory frequency of the observation target, including: based on the target position of the observation target, obtaining a phase time series of the observation target, wherein the phase time series includes phase values ​​of the observation target observed at N observation moments within a specified period time; performing power spectrum estimation on the phase time series to obtain a target power spectrum, and determining the maximum frequency of the target power spectrum as the respiratory frequency of the observation target.

[0046] In order to obtain the respiratory frequency of the observed target, after determining the target position of the observed target, the respiratory frequency of the observed target can be determined by power spectrum estimation. The method of determining the respiratory frequency of the observed target by power spectrum estimation can be: based on the target position of the observed target, obtain the phase time series of the observed target; perform power spectrum estimation on the above phase time series to obtain the target power spectrum; and determine the maximum frequency of the target power spectrum as the respiratory frequency of the observed target.

[0047] Here, the phase time series includes the phase values ​​of the observed target observed at N observation moments within a specified cycle time, wherein the N observation moments are N consecutive observation moments. N is a positive integer greater than or equal to 2, and the value of N can be set based on experience, for example, 200, 300 or other values, which is not limited in this embodiment. In addition, the total duration of the N observation moments is the processing cycle T. For example, when N is 200 and the frequency of obtaining the phase value of the observed target is 20 Hz (that is, the phase value of the observed target is obtained 20 times per second), the processing cycle T is 10 seconds. The larger N is, the longer the processing cycle T is, and the more accurate the target frequency spectrum is, but the system response speed will be reduced accordingly. The value of N can be set according to the requirements of detection accuracy and response speed in actual applications.

[0048] Power spectrum estimation is an important technology in signal processing. By using the N data points that have been obtained to obtain an inaccurate power spectrum, the power spectrum of a real random signal is estimated, which can be used to estimate the power distribution of the signal in the frequency domain. In this embodiment, the power spectrum estimation is mainly used to determine the breathing frequency of the observed target. The power spectrum estimation method in this embodiment may include but is not limited to the periodogram method, the autocorrelation method, the parameter modeling method or the Welch method.

[0049] Through this embodiment, the respiratory frequency of the observed target is determined based on the phase value of the observed target observed at multiple observation times to obtain a phase time series. By performing power spectrum estimation on the phase time series, the phase time series is converted to the frequency domain, and the maximum frequency of the obtained power spectrum is determined as the respiratory frequency of the observed target. This can improve the convenience and accuracy of determining the respiratory frequency.

[0050] In an exemplary embodiment, based on the target position of the observed target, a phase time series of the observed target is obtained, including: based on the target position of the observed target, extracting the target data of the observed target from the observation data of the interior of the vehicle observed by the on-board radar at N observation times, to obtain the target data of the observed target at N observation times; taking the phase of the target data of the observed target at each of the N observation times to obtain a phase time series.

[0051] In this embodiment, in order to obtain the phase time series of the observed target, the target data of the observed target can be first extracted from the observation data of the vehicle interior observed by the vehicle-mounted radar at N observation times (i.e., N observation data) based on the target position of the observed target. The N observation data may include the observation data used to determine the existence of the observed target inside the vehicle, or may be the observation data newly acquired after the existence of the observed target is determined, which is not limited in this embodiment. Since the observation data obtained from one observation includes all information within the observation range of the vehicle-mounted radar, in order to analyze the observed target, the target data of the observed target is intercepted from the observation data of the vehicle-mounted radar based on the target position of the observed target.

[0052] By taking the phase of the target data of the observed target at each of the N observation moments, the above-mentioned phase time series can be obtained. Here, the phase time series is used to describe the change of the phase of the observed target over time, and the change information of the phase of the observed target can be used to describe the change of the distance of the observed target. The phase change information can be used to map the micro-motion of the chest surface of the observed target, and then the breathing frequency value of the target is calculated by estimating the period of the micro-motion. The target data of the observed target returned by the vehicle-mounted radar after detecting the object is represented by a complex signal, which is divided into a real part and an imaginary part. The way to take the phase of the target data of the observed target at each observation moment can be: the real and imaginary data of the target signal are used to calculate the inverse tangent value using a mathematical formula to obtain the phase of the target data.

[0053] For example, the phase data of the target at an observation time is shown in formula (1):

[0054]

[0055] Among them, data1(r, d, 1) is the distance dimension and speed dimension information data obtained by millimeter wave radar, where r is the number of distance units, d is the number of speed units, and phase][indicates the phase of the data.

[0056] The processing period T contains N observation moments, T = [t0, t1, ..., t N-1], and sequentially obtain the phase values ​​of the observed targets to form a phase time series x(n) of the observed targets, where n=0, 1, ..., N-1.

[0057] Through this embodiment, by extracting target data from observation data at multiple observation times and taking phases to form a phase time series, the accuracy of obtaining the time series of the target phase can be improved.

[0058] In an exemplary embodiment, power spectrum estimation is performed on a phase time series to obtain a target power spectrum, including: dividing the phase time series into L segments of data; using a specified window function to convert each segment of data into a corresponding periodogram to obtain a periodogram corresponding to each segment of data; averaging the power spectrum density of the same frequency point in the periodogram corresponding to each segment of data to obtain a target power spectrum.

[0059] In this embodiment, in order to obtain the target power spectrum, the Welch method can be used to estimate the power spectrum of the phase time series to improve the detection efficiency: first, the phase time series is divided into L segments of data, each segment of the L segments of data includes M data in the phase time series, L and M are both positive integers greater than or equal to 2, and the two adjacent segments of data in the L segments of data partially overlap. Here, the overlapping segmentation method is used to divide the data, which can reduce the spectrum leakage introduced by the window function. In addition, the values ​​of L and M can be set according to experience, for example, L = 5, M = 160 or other positive integers greater than or equal to 2. The values ​​of L and M will affect the overlapping part between the two adjacent segments of data. Increasing the overlap between segments can reduce spectrum leakage, and reducing the overlap between segments can reduce the amount of data that is repeatedly calculated, thereby reducing the demand for computing resources. The values ​​of L and M can be set according to the requirements of computing efficiency and computing accuracy in actual applications, and this is not limited in this embodiment.

[0060] Here, Welch power spectrum estimation is a non-parametric method applied in the field of signal processing. It is used to estimate the power spectrum density of the signal and has good computational efficiency and estimation accuracy. The Welch method is also known as the modified periodogram averaging method. Its main idea is: first, divide the long signal into multiple short segments, and perform Fourier transform (FFT) on each short segment; then, average the spectrum of all short segments to obtain the power spectrum density of the entire signal. The Welch method allows for rapid processing and analysis of radar signals, allowing the system to respond in real time and make quick judgments, thereby improving detection efficiency. In addition, while ensuring high accuracy, the Welch method can also reduce the demand for computing resources and reduce system costs.

[0061] For example, after obtaining the aforementioned phase time sequence x(n), the aforementioned phase time sequence x(n) can be divided into L segments of data, wherein each segment of the L segments of data includes M data in the phase time sequence x(n), L and M are both positive integers greater than or equal to 2, and two adjacent segments of data in the L segments of data partially overlap. The i-th segment (i=1, 2, ..., L) of the data after division can be shown as formula (2):

[0062] x i (n)=x(n+(i-1)(Mm)),0≤n≤M-1,1≤i≤L (2)

[0063] The value of n+(i-1)(Mm) is in a constant range, and the value of m can adjust the overlapping portion of two adjacent data segments and satisfies 1≤m≤M-1. For example, if M=5, when m=1, one of the two adjacent data segments overlaps, and the first segment data x1(n) includes x(0) to x(4), that is, the 1st data to the 5th data in the phase time sequence x(n); the second segment data x2(n) includes x(4) to x(8), that is, the 5th data to the 9th data in the phase time sequence x(n), and x(4) is overlapping data; when m=2, two of the two adjacent data segments overlap, and the first segment data x1(n) includes x(0) to x(4), that is, the 1st data to the 5th data in the phase time sequence x(n); the second segment data x2(n) includes x(3) to x(7), that is, the 4th data to the 8th data in the phase time sequence x(n), and x(3) and x(4) are overlapping data. It should be noted that after determining the value of m according to the requirements of the overlapping part between the two segments of data, the values ​​of L and M can be determined according to the value of m, so that the last data x in the last segment of data after division L (M-1) is consistent with the last data in the phase time series x(n).

[0064] After the phase time series x(n) is divided into L segments of data, each segment of data can be converted into a corresponding periodogram using a specified window function to obtain a periodogram corresponding to each segment of data. Among them, the type of the specified window function can be one or more, and can include but is not limited to at least one of the following: Hamming window, rectangular window, Hamming window, etc., which is not limited in this embodiment. Optionally, the specified window function can use a Hamming window to reduce spectrum leakage; or, other window functions with lower complexity can be used to improve calculation efficiency. The type of window function can be selected according to the requirements of calculation efficiency and calculation accuracy in practical applications.

[0065] For example, for the i-th segment of data mentioned above, calculate the periodogram I of the i-th segment of data i (w), as shown in formula (3):

[0066]

[0067] Among them, U is the normalization factor, w(n) is the window function, and here we take a Hamming window with a length of M points.

[0068] The above normalization factor U is shown in formula (4):

[0069]

[0070] After obtaining the periodogram corresponding to each data segment, the Welch power spectrum estimation can be performed. The power spectrum density of the same frequency point in the periodogram corresponding to each data segment can be averaged to obtain the target power spectrum P(w). Here, the average processing can be shown as formula (5):

[0071]

[0072] Finally, the maximum frequency of the target power spectrum P(w) can be determined as the breathing frequency f of the observed target. m . Determine the breathing frequency f of the observed target m The method can be shown as formula (6):

[0073] f m =argmax[P(w)] (6)

[0074] Through this embodiment, the radar signal is subjected to detailed frequency domain analysis through the Welch method, which can effectively identify the tiny vital sign signals left by the target (for example, a child) in the car, thereby improving the accuracy of detection; the power spectral density estimation of the Welch method allows rapid processing and analysis of radar signals, so that the system can respond in real time and make quick judgments, thereby improving detection efficiency; compared with other complex detection systems, the adopted Welch method reduces the demand for computing resources and reduces system costs while ensuring high accuracy.

[0075] In an exemplary embodiment, the in-vehicle data of the current vehicle acquired through an observation by the vehicle-mounted radar is a frame of target observation matrix, and each frame of target observation matrix observed by the vehicle-mounted radar is used to record the range-Doppler matrix observed at the corresponding observation time.

[0076] In order to record the distance and speed information of the observed target, the in-vehicle data of the current vehicle obtained by the vehicle-mounted radar during an observation is recorded in the form of a frame of target observation matrix. Each frame of the target observation matrix can be used to record the range-Doppler matrix observed at the corresponding observation time. The range-Doppler matrix is ​​a key technology for target detection and speed measurement in millimeter-wave radar systems. By analyzing the distance and speed information of the target, accurate positioning and speed estimation of the target can be achieved.

[0077] For example, the target observation matrix is ​​shown in formula (7):

[0078] data1(r,d) (7)

[0079] Among them, r is the distance unit and d is the speed unit.

[0080] Correspondingly, in order to determine the target position of the observed target, a peak search can be performed on the above-mentioned target observation matrix data1 (r, d), including: performing a peak search on the multi-frame target observation matrix observed by the vehicle-mounted radar according to a specified distance interval and a specified speed interval, and determining the position corresponding to the searched peak as the position of an observed target.

[0081] Here, the specified distance interval and the specified speed interval can be specified according to the structure of the vehicle. For example, considering the setting position of the vehicle-mounted radar inside the current vehicle, the length and width of the current vehicle, etc., the distance interval to be searched is specified, and the specified distance interval can cover the interior space of the current vehicle or at least cover all seating areas. For another example, considering the vehicle sizes of different vehicles, the distance interval to be searched is specified, and the specified distance interval can cover the interior space of known vehicles or at least cover all seating areas. Specifically, the specified distance interval can be 0 meters to 2 meters, and its value can be adjusted accordingly according to the size of the interior space. For another example, considering that the vehicle's moving space is limited and the movement ability of the specified target is limited, the specified speed interval can be a speed interval less than the specified speed threshold, or it can be directly zero, that is, searching for targets with zero speed.

[0082] For example, when the specified distance interval is [r1, r2] and the specified speed interval is [d1, d2], the data of the specified distance interval [r1, r2] and speed interval [d1, d2] can be searched for peak values ​​to form distance accumulation data [r m , d m Here, the peak search forms the distance accumulation data [r m , d m ]As shown in formula (8):

[0083] [r m , d m] = arg r，d max[data1(r,d)] (8)

[0084] Among them, max is the operation of calculating the maximum value. After this, the distance accumulation data [r m , d m ] is determined as the location of the observed target.

[0085] After getting the distance accumulation data [r m , d m ], the position corresponding to the searched peak can be determined as the position of an observation target. Here, the distance accumulation data [r m , d m ] where m is the number of peaks found. The number of peaks equals the number of observation targets found.

[0086] Through this embodiment, by performing peak search on the in-vehicle data acquired by the vehicle-mounted radar, the location of the observed target can be accurately determined, thereby improving the accuracy of in-vehicle target detection.

[0087] In an exemplary embodiment, the vehicle-mounted radar may be a vehicle-mounted millimeter-wave radar or other laser radar capable of detecting targets in the vehicle. Compared with other laser radars, the vehicle-mounted millimeter-wave radar has the characteristics of small size, low cost, adaptability to different weather conditions, and can work all day long.

[0088] The interior of the current vehicle is continuously observed through the vehicle-mounted radar, including: within a radar data frame period of the vehicle-mounted radar, the interior of the vehicle is observed through the vehicle-mounted radar, and P receiving channel data of the vehicle-mounted radar are obtained; each receiving channel data of the P receiving channel data is respectively subjected to a two-dimensional fast Fourier transform to obtain P distance-speed information matrices; and the P distance-speed information matrices are subjected to channel accumulation to obtain a frame of target observation matrix.

[0089] In order to obtain the target observation matrix data1 (r, d), the data continuously acquired by the millimeter-wave radar can be channel accumulated. Within a radar data frame period of the vehicle-mounted radar, the interior of the vehicle is observed by the vehicle-mounted radar, and P receiving channel data of the vehicle-mounted radar are obtained, where P is a positive integer greater than or equal to 2, and the P receiving channel data include environmental detection information. Here, the value of P is related to the vehicle-mounted radar used, which can be all or part of the receiving channels of the vehicle-mounted radar. For example, P can be 16, 18 or other values, and its value can be related to the number of receiving channels of the vehicle-mounted radar and the accuracy of the information required to be obtained, etc., which is not limited in this embodiment.

[0090] After obtaining P receiving channel data of the vehicle-mounted radar, a two-dimensional fast Fourier transform (2D-FFT) can be performed on each of the P receiving channel data to obtain P distance-speed information matrices. The above P receiving channel data all belong to data obtained in the same frame. An example of the distance-speed information matrix is ​​shown in formula (9):

[0091] data2(r,d,c) (9)

[0092] Among them, r is the number of distance units, d is the number of velocity units, and c is the number of receiving channels.

[0093] By performing channel accumulation on the above P distance-speed information matrices, a frame target observation matrix data1(r, d) can be obtained, where channel accumulation refers to the accumulation of elements at the same position in the P distance-speed information matrices. Here, the method of performing channel accumulation on the distance-speed information matrix data2(r, d, c) to form the target observation matrix data1(r, d) can be shown in formula (10):

[0094]

[0095] Wherein, abs[] represents the modulus of the data, r is the distance unit, d is the velocity unit, and the target observation matrix data1(r, d) is the range-Doppler matrix.

[0096] Through this embodiment, by means of data collection and channel accumulation, a target observation matrix can be obtained for determining the target position, thereby improving the accuracy of the target position determination.

[0097] In an exemplary embodiment, when the breathing frequency of the observed target is greater than or equal to the specified breathing frequency, a legacy alarm is issued to the observed target, including: when the breathing frequency of the observed target is greater than or equal to the specified breathing frequency, when the distance between the target object bound to the current vehicle and the current vehicle is within a specified distance range, a legacy alarm is issued in at least one of the following ways: controlling an indicator light disposed on the outside of the current vehicle to flash at a specified frequency to issue a legacy visual alarm, wherein when the indicator light flashes, the color of the indicator light is a specified color, or a plurality of rotating colors; controlling a display screen disposed on the outside of the current vehicle to display a specified graphic or a specified symbol to issue a legacy visual alarm; playing an alarm sound through a sound playing component on the current vehicle to issue a legacy auditory alarm; when the target object bound to the current vehicle is not detected within the specified distance range of the current vehicle, sending an alarm message to the terminal device of the target object to remotely alarm the observed target.

[0098] When the breathing rate of the observed target is greater than or equal to the specified breathing rate, it can be determined that a specified target is left behind in the current vehicle (for example, a child is left behind). Here, the specified breathing rate is a breathing rate threshold value calculated and set in advance. In order to remind relevant personnel when it is detected that a specified target is left behind in the vehicle, a legacy alarm can be issued to the surrounding people through the sound and light equipment of the current vehicle. The legacy alarm method includes but is not limited to at least one of the following: controlling the indicator light set on the outside of the current vehicle to flash at a specified frequency to issue a legacy visual alarm; controlling the display screen set on the outside of the current vehicle to display a specified graphic or a specified symbol to issue a legacy visual alarm; playing an alarm sound through the sound playing component on the current vehicle to issue a legacy auditory alarm.

[0099] In addition to the visual and auditory warnings around the current vehicle, a remote warning can also be issued to the target object. The target object is at least one object that is pre-bound to the current vehicle, such as the driver or the driver's relative, etc. The information of the target object's mobile terminal device can be recorded on the current vehicle. The above-mentioned mobile terminal device can be a mobile phone, and the information of the mobile terminal device can be a mobile phone number. Taking the terminal device of the target object as a mobile phone as an example, when it is determined that a designated target is left in the current vehicle, and the target object bound to the current vehicle is not detected within the specified distance range of the current vehicle (that is, the target object is not around the current vehicle), an alarm message can be sent to the target object through the recorded mobile phone number (for example, sending a text message or initiating a call) for remote warning.

[0100] In particular, the above-mentioned visual warning, auditory warning and remote warning methods can be executed individually or in combination. For example, if it is determined that a child is left in the current vehicle and the driver or the driver's relatives are not around the current vehicle, the indicator light on the outside of the current vehicle can be controlled to flash, and the current vehicle can be controlled to sound the horn, and a text message can be sent to the target object's mobile phone to achieve a combined warning.

[0101] In addition, different warning methods can be used according to the length of the car lock time, the current time, the current temperature, etc. For example, when the car lock time is long and the current temperature is high, the combined warning method can be used first, and the volume of the warning sound can be increased, or the car lock can be automatically opened to ensure the safety of children left in the car. The warning method and conditions can be flexibly set according to experience, and this is not limited in this embodiment.

[0102] Through this embodiment, when it is detected that a designated target is left behind, an alarm can be issued to the driver through multiple means such as vision and hearing, and a remote notification function is provided, which can enhance the user experience and ensure the safety of the target left behind in the car.

[0103] The following is an explanation of the in-vehicle target detection method in the embodiment of the present application in conjunction with an optional example. In this optional example, the designated target is a child, the vehicle-mounted radar is a vehicle-mounted millimeter-wave radar, the target observation matrix is ​​data1, the distance-speed information matrix is ​​data2, the statistical period is T, and the statistical time of the vehicle-mounted radar obtaining information within the statistical period T is t.

[0104] This optional example provides a vehicle-mounted millimeter-wave radar child abandonment detection method based on the Welch method. By comparing the statistical target frequency with the child's breathing frequency f within the processing cycle T, if it exceeds the judgment threshold of the breathing signal, it is determined that a child is present and a child abandonment alarm is issued. The process of this method can be as follows Figure 4 As shown, the following steps are included: step S402, data collection; step S404, target position estimation S404; step S406, target spectrum estimation; step S408, child left-behind detection.

[0105] Combine the following Figure 5 The implementation process of the vehicle-mounted millimeter-wave radar child left behind detection method in this optional example is explained. Figure 5 is a flow chart of another optional in-vehicle target detection method according to an embodiment of the present application, such as Figure 5 As shown, the process of the method mainly includes steps S502 to S522.

[0106] Step S502, the current vehicle is locked and in-vehicle object detection begins;

[0107] Step S504, the vehicle-mounted radar acquires a frame of data, and performs 2D Fourier transform on the data to obtain data1;

[0108] Step S506, the channel accumulates data to obtain data2.

[0109] The aforementioned data collection (step S402) may include the aforementioned steps S502 to S504.

[0110] Step S508, perform peak search on data2 to estimate the target position.

[0111] The aforementioned target position estimation (step S404) may include the aforementioned step S508.

[0112] Step S510, determining whether the statistical time t is greater than the statistical period T, if it is satisfied, executing step S512, if not, returning to executing step S504;

[0113] Step S512, performing time series analysis on the data to obtain phase data;

[0114] Step S514: perform Welch spectrum estimation on the phase data to obtain the respiratory frequency of the observed target.

[0115] The aforementioned target spectrum estimation (ie, step S406 ) may include the aforementioned steps S510 to S514 .

[0116] Step S516, determining whether the target breathing rate is greater than the child's breathing rate, if yes, executing step S518, if not, executing step S520;

[0117] Step S518, determining that a child is left behind, giving an alarm, and executing step S522;

[0118] Step S520: determine that no child is left behind, and execute step S522.

[0119] The aforementioned child left behind detection (step S408) includes the aforementioned steps S516 to S520.

[0120] Step S522, end the detection of this statistical cycle and enter the next statistical cycle.

[0121] Through this optional example, the vehicle-mounted radar child abandonment detection method based on the Welch method can accurately and efficiently detect whether a child is left behind in the vehicle, and it is user-friendly, has a low system setting cost, and has a simple implementation process and strong practicality.

[0122] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0123] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM (Read-Only Memory) / RAM (Random Access Memory), a disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0124] According to another aspect of the embodiments of the present application, there is also provided an in-vehicle target detection device, which is used to implement the in-vehicle target detection method provided in the above-mentioned embodiments, and will not be repeated hereafter. As used below, the term "module" may be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0125] Figure 6 is a structural block diagram of an optional in-vehicle target detection device according to an embodiment of the present application, such as Figure 6 As shown in , the device includes: an observation unit 602, a detection unit 604 and an alarm unit 606.

[0126] The observation unit 602 is used to continuously observe the interior of the current vehicle through the vehicle-mounted radar when the current vehicle is in a locked state, wherein the observation range of the vehicle-mounted radar covers the seat area of ​​the current vehicle. The observation unit 602 can be used to execute step S202.

[0127] The detection unit 604 is used to detect the breathing frequency of the observation target based on the target position of the observation target when it is determined that there is an observation target inside the vehicle according to the observation data inside the vehicle, so as to obtain the breathing frequency of the observation target. The detection unit 604 can be used to execute step S204.

[0128] The alarm unit 606 is used to issue a legacy alarm to the observed target when the respiratory frequency of the observed target is greater than or equal to the specified respiratory frequency. The alarm unit 606 can be used to execute step S206.

[0129] Through the embodiments provided by the present application, when the current vehicle is in a locked state, the interior of the current vehicle is continuously observed through the on-board radar, wherein the observation range of the on-board radar covers the seat area of ​​the current vehicle; when it is determined that there is an observation target inside the vehicle based on the observation data inside the vehicle, the breathing rate of the observation target is detected based on the target position of the observation target to obtain the breathing rate of the observation target; when the breathing rate of the observation target is greater than or equal to the specified breathing rate, a legacy alarm is issued for the observation target, thereby solving the technical problem of poor target detection accuracy in the in-vehicle target detection method in the related art and improving the accuracy of in-vehicle target detection.

[0130] In an exemplary embodiment, the detection unit includes: an acquisition module, which is used to acquire a phase time series of the observed target based on the target position of the observed target, wherein the phase time series includes the phase values ​​of the observed target observed at N observation moments within a specified period time; a first execution module, which is used to perform power spectrum estimation on the phase time series to obtain a target power spectrum, and determine the maximum frequency of the target power spectrum as the breathing frequency of the observed target.

[0131] In an exemplary embodiment, the acquisition module includes: a first extraction submodule, which is used to extract the target data of the observed target from the observation data of the interior of the vehicle observed by the vehicle-mounted radar at N observation times based on the target position of the observed target, and obtain the target data of the observed target at N observation times; a second extraction submodule, which is used to take the phase of the target data of the observed target at each of the N observation times to obtain a phase time series.

[0132] In an exemplary embodiment, the first execution module includes: a division submodule, which is used to divide the phase time series into L segments of data, wherein each segment of the L segments of data includes M data in the phase time series, L and M are both positive integers greater than or equal to 2, and two adjacent segments of the L segments of data partially overlap; a conversion submodule, which is used to use a specified window function to convert each segment of data into a corresponding periodogram, so as to obtain a periodogram corresponding to each segment of data; and a processing submodule, which is used to average the power spectral density of the same frequency point in the periodogram corresponding to each segment of data, so as to obtain a target power spectrum.

[0133] In an exemplary embodiment, the in-vehicle data of the current vehicle obtained by an observation through a vehicle-mounted radar is a frame of target observation matrix, and each frame of target observation matrix observed by the vehicle-mounted radar is used to record the distance-Doppler matrix observed at the corresponding observation time; the above-mentioned device also includes: an execution unit, which is used to perform peak search on multiple frames of target observation matrices observed by the vehicle-mounted radar according to a specified distance interval and a specified speed interval, and determine the position corresponding to the searched peak as the position of an observed target.

[0134] In an exemplary embodiment, the vehicle-mounted radar is a vehicle-mounted millimeter-wave radar; the observation unit includes: a second execution module, which is used to observe the interior of the vehicle through the vehicle-mounted radar within a radar data frame period of the vehicle-mounted radar, and obtain P receiving channel data of the vehicle-mounted radar, wherein P is a positive integer greater than or equal to 2, and the P receiving channel data include environmental detection information; a transformation module, which is used to perform a two-dimensional fast Fourier transform on each of the P receiving channel data to obtain P distance-speed information matrices; an accumulation module, which is used to perform channel accumulation on the P distance-speed information matrices to obtain a frame of target observation matrix, wherein channel accumulation refers to the accumulation of elements at the same position in the P distance-speed information matrices.

[0135] In an exemplary embodiment, the alarm unit includes: an alarm module, which is used to issue a legacy alarm in at least one of the following ways when the breathing frequency of the observed target is greater than or equal to the specified breathing frequency and when the distance between the target object bound to the current vehicle and the current vehicle is within a specified distance range: controlling an indicator light arranged on the outside of the current vehicle to flash at a specified frequency to issue a legacy visual alarm, wherein when the indicator light flashes, the color of the indicator light is a specified color, or a plurality of rotating colors; controlling a display screen arranged on the outside of the current vehicle to display a specified graphic or a specified symbol to issue a legacy visual alarm; playing an alarm sound through a sound playing component on the current vehicle to issue a legacy auditory alarm; a sending module, which is used to send an alarm message to a terminal device of the target object to remotely alarm the observed target when the target object bound to the current vehicle is not detected within the specified distance range of the current vehicle.

[0136] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein the program executes the steps of any of the above method embodiments when it is run.

[0137] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a ROM, a RAM, a mobile hard disk, a magnetic disk, or an optical disk.

[0138] According to another aspect of the embodiments of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to execute the steps in any of the above method embodiments through the computer program. In an exemplary embodiment, the electronic device may further comprise a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0139] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0140] According to another aspect of the embodiment of the present application, a computer program product is also provided, which includes a computer program / instruction, and the computer program / instruction contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 706, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit 701, various functions provided by the embodiment of the present application are executed. The above-mentioned serial numbers of the embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0141] Figure 7 The computer system structure block diagram of the electronic device used to implement the embodiment of the present application is schematically shown. Figure 7 As shown, the computer system 700 includes a CPU (Central Processing Unit) 701, which can perform various appropriate actions and processes according to the program stored in the ROM 702 or the program loaded from the storage part 708 to the RAM 703. Various programs and data required for system operation are also stored in the random access memory 703. The central processing unit 701, the read-only memory 702 and the random access memory 703 are connected to each other through a bus 704. An I / O (Input / Output) interface 705 is also connected to the bus 704.

[0142] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a CRT (Cathode Ray Tube), an LCD (Liquid Crystal Display), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 706 including a network interface card such as a LAN card, a modem, etc. The communication section 706 performs communication processing via a network such as the Internet. A drive 710 is also connected to the input / output interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed so that a computer program read therefrom is installed into the storage section 708 as needed.

[0143] In particular, according to an embodiment of the present application, the process described in each method flow chart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program contains a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 706, and / or installed from the removable medium 711. When the computer program is executed by the central processor 701, various functions defined in the system of the present application are executed.

[0144] It should be noted that Figure 7 The computer system 700 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0145] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0146] The above are only preferred embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting an in-vehicle target, characterized in that: include: When the current vehicle is in a locked state, continuously observing the interior of the current vehicle through a vehicle-mounted radar, wherein the observation range of the vehicle-mounted radar covers the seat area of ​​the current vehicle; When it is determined that an observation target exists inside the vehicle according to the observation data inside the vehicle, based on the target position of the observation target, a breathing frequency detection is performed on the observation target to obtain the breathing frequency of the observation target; When the respiratory frequency of the observed target is greater than or equal to the specified respiratory frequency, a legacy alarm is issued to the observed target.

2. The method according to claim 1, characterized in that The step of detecting the respiratory frequency of the observation target based on the target position of the observation target to obtain the respiratory frequency of the observation target includes: Based on the target position of the observed target, a phase time series of the observed target is obtained, wherein the phase time series includes phase values ​​of the observed target observed at N observation moments within a specified period time; A power spectrum is estimated for the phase time series to obtain a target power spectrum, and the maximum frequency of the target power spectrum is determined as the breathing frequency of the observed target.

3. The method according to claim 2, characterized in that The step of acquiring a phase time series of the observed target based on the target position of the observed target comprises: Based on the target position of the observation target, extracting the target data of the observation target from the observation data of the interior of the vehicle observed by the vehicle-mounted radar at the N observation times, to obtain the target data of the observation target at the N observation times; The target data of the observed target at each of the N observation moments is phased to obtain the phase time series.

4. The method according to claim 2, characterized in that: The performing power spectrum estimation on the phase time series to obtain a target power spectrum includes: Divide the phase time series into L segments of data, wherein each segment of the L segments of data includes M data in the phase time series, L and M are both positive integers greater than or equal to 2, and two adjacent segments of data in the L segments of data partially overlap; Using a specified window function to convert each segment of data into a corresponding periodogram, to obtain a periodogram corresponding to each segment of data; The power spectrum density of the same frequency point in the periodogram corresponding to each segment of data is averaged to obtain the target power spectrum.

5. The method according to claim 1, characterized in that The in-vehicle data in the current vehicle obtained by performing one observation by the vehicle-mounted radar is a frame of target observation matrix, and each frame of the target observation matrix observed by the vehicle-mounted radar is used to record the range-Doppler matrix observed at the corresponding observation time; the method also includes: A peak search is performed on the target observation matrix of multiple frames observed by the vehicle-mounted radar according to a specified distance interval and a specified speed interval, and a position corresponding to the searched peak is determined as a position where the observed target is located.

6. The method according to claim 5, characterized in that The vehicle-mounted radar is a vehicle-mounted millimeter-wave radar; and the continuously observing the interior of the current vehicle by the vehicle-mounted radar includes: In a radar data frame period of the vehicle-mounted radar, the interior of the vehicle is observed by the vehicle-mounted radar, and P receiving channel data of the vehicle-mounted radar are obtained, wherein P is a positive integer greater than or equal to 2, and the P receiving channel data include environment detection information; Performing a two-dimensional fast Fourier transform on each of the P receiving channel data to obtain P distance-speed information matrices; Channel accumulation is performed on the P distance-speed information matrices to obtain a frame of the target observation matrix, wherein the channel accumulation refers to the accumulation of elements at the same position in the P distance-speed information matrices.

7. The method according to any one of claims 1 to 6, characterized in that When the respiratory frequency of the observed target is greater than or equal to the specified respiratory frequency, issuing a legacy alarm to the observed target includes: In the case where the breathing frequency of the observed target is greater than or equal to the specified breathing frequency, when the distance between the target object bound to the current vehicle and the current vehicle is within the specified distance range, a legacy alarm is performed in at least one of the following ways: controlling an indicator light arranged on the outside of the current vehicle to flash at a specified frequency to perform a legacy visual alarm, wherein when the indicator light flashes, the color of the indicator light is a specified color, or a plurality of colors that rotate in rotation; controlling a display screen arranged on the outside of the current vehicle to display a specified graphic or a specified symbol to perform a legacy visual alarm; playing an alarm sound through a sound playing component on the current vehicle to perform a legacy auditory alarm; When the target object bound to the current vehicle is not detected within the specified distance range of the current vehicle, an alarm message is sent to the terminal device of the target object to remotely alarm the observed target.

8. An in-vehicle target detection device, characterized in that: include: an observation unit, configured to continuously observe the interior of the current vehicle through a vehicle-mounted radar when the current vehicle is in a locked state, wherein the observation range of the vehicle-mounted radar covers the seat area of ​​the current vehicle; a detection unit, configured to, when it is determined that an observation target exists inside the vehicle according to the observation data inside the vehicle, detect the breathing frequency of the observation target based on the target position of the observation target to obtain the breathing frequency of the observation target; The alarm unit is used to issue a legacy alarm to the observation target when the breathing frequency of the observation target is greater than or equal to the specified breathing frequency.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method described in any one of claims 1 to 7 when executed by a processor.

10. An electronic 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, the steps of the method described in any one of claims 1 to 7 are implemented.

Images