Multi-target vital sign detection method, terminal equipment and storage medium

Abstract

The invention is suitable for the field of data processing, and discloses a multi-target vital sign detection method, terminal equipment and a storage medium. The multi-target vital sign detection method comprises the following steps: controlling a millimeter wave radar to scan a target area to obtain a point cloud sequence; determining the number of target persons in the target area according to the point cloud sequence; when the number of the target persons is multiple, generating an independent point cloud sequence of each target person according to the point cloud sequence; and according to the independent point cloud sequence of each target person, generating a vital sign parameter corresponding to each target person. According to the invention, the hardware complexity and the hardware cost are reduced, and the advantages of non-contact detection of the millimeter wave radar are maintained.

Description

Technical Field

[0001] The present invention belongs to the field of data processing, and in particular relates to a multi-target vital sign detection method, terminal equipment and storage medium. Background Art

[0002] Multi-target detection refers to the simultaneous identification, location, and tracking of multiple independent targets in the same scene. In vital sign detection, multi-target detection capabilities rely heavily on hardware redundancy, and commercial applications face significant cost barriers.

[0003] Traditional solutions require multiple-input, multiple-output antenna arrays or frequency-modulated continuous-wave broadband radars to separate multiple target signals through beamforming. The high hardware cost of a single device makes it difficult to meet the needs of universal access in scenarios like home monitoring and nursing homes. A new technical approach is needed to address these challenges. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a multi-target vital sign detection method, terminal device, and storage medium, which can solve the problem of high hardware cost of a single device in related technologies.

[0005] A first aspect of the present invention provides a multi-target vital sign detection method, comprising: Control the millimeter-wave radar to scan the target area and obtain a point cloud sequence; determining the number of target persons in the target area according to the point cloud sequence; When there are multiple target persons, generating an independent point cloud sequence for each target person according to the point cloud sequence; According to the independent point cloud sequence of each target person, the vital sign parameters corresponding to each target person are generated.

[0006] Optionally, in a first implementation of the first aspect of the present invention, the step of generating a vital sign parameter corresponding to each target person according to the independent point cloud sequence of each target person includes: extracting point cloud position change information corresponding to tiny physiological movements from the independent point cloud sequence of each target person; Processing the position change information of each point cloud to separate time-varying signals representing different vital signs, wherein the time-varying signals include respiratory motion signals and cardiac motion signals; Each of the time-varying signals is analyzed to obtain vital sign parameters corresponding to each of the target persons, wherein the vital sign parameters include respiratory rate and heart rate parameters.

[0007] Optionally, in a second implementation of the first aspect of the present invention, before the step of processing the position change information of each point cloud to separate the time-varying signals representing different vital signs, the method further includes: Dynamic interference filtering is performed on each point cloud position change information, where the dynamic interference filtering includes identifying and suppressing noise signals generated by non-physiological movements of the target person.

[0008] Optionally, in a third implementation of the first aspect of the present invention, the step of determining the number of target persons in the target area according to the point cloud sequence includes: Analyzing the distribution characteristics of the point cloud sequence in the spatial domain; identifying point cloud clusters representing different potential persons based on the distribution characteristics; The number of target point cloud clusters that persist in a continuous time window and meet the characteristics of the target person is counted to obtain the number of the target persons in the target area.

[0009] Optionally, in a fourth implementation of the first aspect of the present invention, when there are multiple target persons, the step of generating an independent point cloud sequence for each target person according to the point cloud sequence includes: Based on the point cloud sequence, identifying and tracking spatially separated features of multiple targets; constructing an independent motion trajectory for each target person according to the spatial separation feature; Based on the motion trajectory, the point cloud data belonging to different target persons in the point cloud sequence are separated to obtain an independent point cloud sequence corresponding to each target person.

[0010] Optionally, in a fifth implementation of the first aspect of the present invention, the step of controlling the millimeter-wave radar to scan the target area to obtain a point cloud sequence includes: Control the millimeter-wave radar to scan the target area and obtain the original point cloud sequence; The original point cloud sequence is subjected to multi-dimensional dynamic enhancement processing to obtain the point cloud sequence, wherein the multi-dimensional dynamic enhancement processing includes Doppler dimension filtering, spatial domain motion amplification and / or time domain signal focusing.

[0011] Optionally, in a sixth implementation of the first aspect of the present invention, when there are multiple target persons, after the step of generating an independent point cloud sequence for each target person according to the point cloud sequence, the method further includes: When a spatially overlapping independent point cloud sequence group appears in the plurality of independent point cloud sequences, a multi-target trajectory decoupling process is performed on the independent point cloud sequence group.

[0012] Optionally, in a seventh implementation manner of the first aspect of the present invention, the step of performing multi-target trajectory decoupling processing on the independent point cloud sequence group includes: extracting differentiated motion patterns from mixed signals corresponding to the independent point cloud sequence groups; establishing a motion separation model based on the differentiated motion patterns; The independent point cloud sequence group is processed according to the motion separation model to obtain a decoupled independent point cloud sequence output by the motion separation model.

[0013] In a second aspect, an embodiment of the present invention provides a terminal device 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 the above-mentioned multi-target vital signs detection method when executing the computer program.

[0014] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the above-mentioned multi-target vital signs detection method.

[0015] In a fourth aspect, an embodiment of the present invention provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute the multi-target vital sign detection method.

[0016] Compared with the prior art, the beneficial effects of the embodiments of the present invention are as follows: through the point cloud sequence collected by millimeter-wave radar, the separation and tracking of multiple targets are achieved through steps such as target number identification and independent point cloud sequence generation at the data processing layer, eliminating the reliance on expensive hardware redundancy. Traditional solutions rely on hardware beamforming technology to separate multi-target signals, resulting in high equipment costs; the present invention replaces complex RF hardware with a post-processing algorithm for the point cloud sequence, and only requires a single device standard configuration to complete the spatial resolution and independent extraction of vital signs of multiple targets. This reduces hardware complexity and hardware costs while maintaining the advantages of non-contact detection of millimeter-wave radar. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A schematic diagram of an embodiment of a multi-target vital sign detection method according to an embodiment of the present invention; Figure 2This is a schematic diagram of a specific embodiment of step S102 of the multi-target vital sign detection method in an embodiment of the present invention; Figure 3 This is a schematic diagram of a specific embodiment of step S102 of the multi-target vital sign detection method in an embodiment of the present invention; Figure 4 This is a schematic diagram of a specific embodiment of step S103 of the multi-target vital sign detection method in an embodiment of the present invention; Figure 5 Schematic diagram of a terminal device in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are protected by the present invention.

[0020] It should be noted that the terms "include", "comprising" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, terminal, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices. In the claims, specification and drawings of the present invention, relational terms such as "first" and "second" are merely used to distinguish one entity / operation / object from another entity / operation / object, and do not necessarily require or imply any such real-time relationship or order between these entities / operations / objects.

[0021] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0022] Multi-target detection refers to the simultaneous identification, location, and tracking of multiple independent targets in the same scene. In vital sign detection, multi-target detection capabilities rely heavily on hardware redundancy, and commercial applications face significant cost barriers.

[0023] Traditional solutions require multiple-input, multiple-output antenna arrays or frequency-modulated continuous-wave broadband radars to separate multiple target signals through beamforming. The high hardware cost of a single device makes it difficult to meet the needs of universal access in scenarios like home monitoring and nursing homes. A new technical approach is needed to address these challenges.

[0024] In view of this, an embodiment of the present invention provides a multi-target vital sign detection method, terminal device and storage medium. Through the point cloud sequence collected by millimeter-wave radar, the separation and tracking of multiple targets are achieved through target quantity identification, independent point cloud sequence generation and other steps at the data processing layer, eliminating the reliance on expensive hardware redundancy. Traditional solutions rely on hardware beamforming technology to separate multi-target signals, resulting in high equipment costs. The present invention replaces complex RF hardware with a post-processing algorithm of the point cloud sequence, and only a single device standard configuration is required to complete the spatial resolution and independent extraction of vital signs of multiple targets. It reduces the hardware complexity and hardware cost, and also maintains the advantages of non-contact detection of millimeter-wave radar.

[0025] In order to illustrate the technical solution of the present invention, specific embodiments are provided below.

[0026] Figure 1 The following is a schematic diagram of a multi-target vital sign detection method according to an embodiment of the present invention. The method can be applied to a terminal device, such as a mobile phone, tablet computer, laptop computer, ultra-mobile personal computer (UMPC), or netbook.

[0027] Specifically, the multi-target vital sign detection method may include the following steps S101 to S104.

[0028] Step S101: Control the millimeter-wave radar to scan the target area to obtain a point cloud sequence.

[0029] In an embodiment of the present invention, the millimeter wave radar is first initialized to drive it to transmit radio frequency signals and receive reflected echoes from the target area.

[0030] The radar beam is controlled to periodically scan the target area, continuously collecting raw sensor data. The raw echo signals are then subjected to three-dimensional calculations of range, angle, and Doppler, generating a dynamically updated point cloud sequence in real time. This point cloud sequence contains multi-dimensional features such as the spatial coordinates, motion speed, and signal strength of all reflectors within the target area, forming the foundational dataset for subsequent processing.

[0031] Step S102: determining the number of target persons in the target area according to the point cloud sequence.

[0032] In an embodiment of the present invention, based on the generated continuous point cloud sequence, its spatiotemporal distribution characteristics in three-dimensional space are analyzed. Clustering algorithms can be used to identify point cloud clusters with human morphological characteristics and verify the dynamic continuity of these point cloud clusters.

[0033] Finally, the number of independent point cloud clusters that meet the "target person" judgment conditions within the set time window is counted, and the total number of target people actually existing in the target area is output.

[0034] Step S103 : When there are multiple target persons, an independent point cloud sequence is generated for each target person according to the point cloud sequence.

[0035] In the embodiment of the present invention, this process is initiated when it is determined that the number of target persons is ≥ 2. First, the spatial topological relationship of the point cloud sequence is analyzed to extract the differentiated features of each target person, including spatial position distribution, motion trajectory independence, velocity vector difference, etc.

[0036] A multi-target tracking algorithm is then used to construct a unique motion trajectory model for each target person. Finally, the original point cloud sequence is partitioned based on the trajectory model, and the point cloud data belonging to different targets is reorganized into independent, time-synchronized point cloud sequence streams.

[0037] Step S104 : generating vital sign parameters corresponding to each target person according to the independent point cloud sequence of each target person.

[0038] In this embodiment of the present invention, physiological signal feature extraction is performed on each independent point cloud sequence output. Signal transformation is used to analyze the micro-motion components of the point cloud position data, separating the physiological modulation signals representing vital signs. Finally, a time-frequency domain joint solution is performed on the respiratory motion frequency and cardiac harmonic components, outputting core vital sign parameters (including quantitative indicators such as respiratory rate and heart rate) for each target person, enabling parallel vital sign detection of multiple targets.

[0039] Compared with the prior art, the beneficial effects of the embodiments of the present invention are as follows: through the point cloud sequence collected by millimeter-wave radar, the separation and tracking of multiple targets are achieved through steps such as target number identification and independent point cloud sequence generation at the data processing layer, eliminating the reliance on expensive hardware redundancy. Traditional solutions rely on hardware beamforming technology to separate multi-target signals, resulting in high equipment costs; the present invention replaces complex RF hardware with a post-processing algorithm for the point cloud sequence, and only requires a single device standard configuration to complete the spatial resolution and independent extraction of vital signs of multiple targets. This reduces hardware complexity and hardware costs while maintaining the advantages of non-contact detection of millimeter-wave radar.

[0040] Traditional millimeter-wave vital sign detection has technical flaws. On the one hand, conventional radar algorithms treat the human body as a whole reflection source, and chest micro-movements are easily drowned out by torso displacement noise. On the other hand, the harmonic components of the respiratory fundamental frequency overlap with the heartbeat fundamental frequency in the frequency domain, making it easy for traditional bandpass filters to fail to separate the effective signal. Traditional solutions only support single-target vital sign detection. In multi-target scenarios, the error rate is high due to signal cross-interference, which seriously restricts the implementation of this technology in scenarios with urgent needs such as elderly care monitoring and multi-person wards. Based on this, the present invention proposes an optional embodiment.

[0041] Reference Figure 2 , Figure 2 This is a schematic diagram of a specific embodiment of step S102 of the multi-target vital sign detection method in an embodiment of the present invention. Step S104 also includes the following specific implementation methods.

[0042] Step S1041 : extracting point cloud position change information corresponding to tiny physiological movements from the independent point cloud sequence of each target person.

[0043] In this embodiment, spatiotemporal analysis is performed on each individual point cloud sequence, focusing on millimeter-level periodic changes in the point cloud coordinates. A motion amplification algorithm is used to enhance physiologically relevant micro-displacement signals, such as chest rise and fall and body pulsation, while simultaneously suppressing interference from ambient noise and static reflectors. The resulting output is point cloud position change information directly linked to vital signs.

[0044] Step S1042 : Process the position change information of each point cloud to separate the time-varying position change information representing different vital signs and input the time-varying position change information extracted in step 1 into a signal separation module.

[0045] In an embodiment of the present invention, the respiratory frequency band and the cardiac frequency band are separated by bandpass filtering, and then independent component analysis (ICA) is performed on the aliased signal, and finally a parallel time domain waveform time-varying signal of a pure respiratory motion signal and a cardiac motion signal is output, and the time-varying signal includes the respiratory motion signal and the cardiac motion signal.

[0046] Step S1043 , analyzing each of the time-varying signals to obtain vital sign parameters corresponding to each of the target persons, wherein the vital sign parameters include respiratory rate and heart rate parameters.

[0047] In an embodiment of the present invention, a peak detection algorithm is applied to the separated respiratory motion signal to calculate the respiratory frequency by counting the number of respiratory cycles per unit time. A wavelet transform is performed on the cardiac motion signal to extract QRS waveform feature points, and heart rate parameters are calculated based on the RR interval to achieve quantitative output of vital sign parameters.

[0048] In this embodiment of the present invention, hierarchical signal processing can significantly improve the reliability of multi-target vital sign detection. First, point cloud micro-motion extraction overcomes the bottleneck of traditional radar's inability to identify sub-centimeter physiological movements. By focusing on the analysis of point cloud position changes, macroscopic motion data can be converted into microscopic vital signs. Second, frequency domain decoupling of respiratory and cardiac signals effectively resolves vital sign aliasing interference, overcoming the problem of traditional single filters causing the heartbeat signal to be submerged by respiratory harmonics. Finally, a dual-modal parallel quantization algorithm eliminates the cross-influence between vital sign parameters, ensuring the independent accuracy of respiratory frequency and heart rate parameters.

[0049] Traditional millimeter-wave vital sign detection faces the dilemma of overwhelming motion interference. Fixed-threshold filtering can misinterpret valid physiological signals. Spectral filtering cannot distinguish interference from vital signs within the same frequency band. Motion compensation algorithms rely on additional inertial sensors, increasing costs. To address this issue, the present invention proposes an alternative embodiment.

[0050] Step S1042 also includes the following specific implementation methods.

[0051] Step S10421 : performing dynamic interference filtering on each point cloud position change information, wherein the dynamic interference filtering includes identifying and suppressing noise signals generated by non-physiological movements of the target person.

[0052] In an embodiment of the present invention, the position change information of each target point cloud is analyzed for motion patterns, and physiological motion and non-physiological motion are distinguished through time domain acceleration detection and frequency domain energy analysis. Data segments that meet the interference characteristics are marked.

[0053] A motion compensation algorithm eliminates baseline drift caused by limb displacement, then applies a threshold filter based on short-term energy entropy to address sudden noise. Finally, a Kalman prediction algorithm corrects the physiological motion trajectory. The output retains purified position change information, retaining pure physiological micro-motion, for subsequent use in vital sign signal separation.

[0054] In this embodiment of the present invention, the identification and suppression of non-physiological motion solves the core pain point of traditional solutions, whereby the vital sign signals are completely ineffective due to slight movements of the target person. The adaptive filter chain can reduce respiratory rate misjudgments caused by motion artifacts; effectively avoid the masking effect of sudden noise on the heartbeat signal; and preserve the phase integrity of the original physiological signal, providing a pure data source for multi-sign separation. This is especially true in scenarios such as elderly care monitoring, where frequent changes in the elderly's body position no longer affect monitoring continuity, significantly improving the reliability of the technology's implementation.

[0055] Traditional millimeter-wave target statistics rely on a single spatial clustering algorithm, which can easily misclassify static objects such as chair backs (1.1 meters in height) and potted plants (with point cloud density similar to that of a human body) as human bodies, resulting in a high false alarm rate. To address this issue, the present invention proposes an alternative embodiment.

[0056] Reference Figure 3 , Figure 3 4 is a schematic diagram of a specific embodiment of step S102 of the multi-target vital sign detection method in an embodiment of the present invention. Step S102 also includes the following specific implementation methods.

[0057] Step S1021: Analyze the distribution characteristics of the point cloud sequence in the spatial domain.

[0058] In this embodiment of the present invention, spatial clustering analysis is performed on the continuous point cloud sequence generated by the millimeter-wave radar. A density clustering algorithm is used to identify high-density point cloud areas in three-dimensional space. Combined with human body geometric feature constraints, potential human target point cloud clusters are initially screened.

[0059] Step S1022: identifying point cloud clusters representing different potential persons based on the distribution characteristics.

[0060] In an embodiment of the present invention, the initially screened point cloud clusters are verified for dynamic features related to vital signs.

[0061] Track the continuity of point cloud clusters within a sliding time window to eliminate transient interference. Detect the energy spectrum of micro-motion signals within the point cloud cluster to confirm the presence of a characteristic spectrum consistent with the respiratory / heartbeat frequency band (0.1-2Hz). Use trajectory prediction to eliminate motion patterns that are inconsistent with the human body.

[0062] Step S1023 , counting the number of target point cloud clusters that exist continuously in the continuous time window and meet the characteristics of the target person, and obtaining the number of the target persons in the target area.

[0063] In an embodiment of the present invention, the independence of point cloud clusters that meet the characteristics of the target person is determined.

[0064] If the horizontal distance between the centroids of two point cloud clusters is continuously greater than 1 meter, they are considered independent individuals.

[0065] Update the spatial position relationship of each target in real time.

[0066] The number of independent point cloud clusters that exist stably in n consecutive scanning cycles is determined as the number of target people.

[0067] In the embodiment of the present invention, target statistics realizes the number of people perception in multiple scenarios. By using the dual feature constraints of space and dynamics, the false alarm rate can be reduced.

[0068] Conventional millimeter wave multi-target separation results in track swapping when targets cross each other. Based on this, the present invention proposes an alternative embodiment.

[0069] Reference Figure 4 , Figure 44 is a schematic diagram of a specific embodiment of step S103 of the multi-target vital sign detection method in an embodiment of the present invention. Step S103 also includes the following specific implementation methods.

[0070] Step S1031 : Based on the point cloud sequence, identifying and tracking the spatial separation features of multiple targets.

[0071] In this embodiment, a multi-target motion state space model is constructed based on real-time point cloud sequences. Spatial separation features are extracted through 3D point cloud clustering and feature vector calculation. Specifically, the centroid distances between potential target point cloud clusters are quantified; the Doppler velocity distributions of different target point cloud clusters are analyzed; and parameters such as the aspect ratio of the point cloud clusters are calculated to verify human morphological characteristics. Ultimately, a multi-target feature fingerprint library is generated, providing a classification basis for trajectory modeling.

[0072] Step S1032: constructing an independent motion trajectory for each target person according to the spatial separation feature.

[0073] In this embodiment of the present invention, a multi-model tracker is activated for each identified target person. Specifically, the target's next frame position is deduced using a Kalman filter; corresponding point cloud clusters in consecutive frames are matched using feature fingerprints; and the motion path is smoothed using a dynamic time warping algorithm. This outputs a stable and spatially isolated chain of individual motion trajectories.

[0074] Step S1033 : Separate the point cloud data belonging to different target persons in the point cloud sequence according to the motion trajectory to obtain an independent point cloud sequence corresponding to each target person.

[0075] In this embodiment of the present invention, a spatiotemporal allocation matrix is established based on the motion trajectory, assigning each point cloud frame in the original point cloud sequence to a specific location. Specifically, the Mahalanobis distance between the point cloud points and the predicted positions of each trajectory is calculated; the temporal consistency of the point cloud along the trajectory chain is checked; and interference points with signal-to-noise ratios below a threshold or cross-trajectory jumps are filtered out, ultimately generating a stream of independent point cloud sequences corresponding to the target person.

[0076] In this embodiment of the present invention, the target separation method can significantly improve the robustness of vital sign detection in complex scenarios. Spatially separated feature recognition can reduce the minimum distinguishable distance, solving the problem of traditional solutions failing in scenarios such as multiple people walking side by side or staggered.

[0077] Traditional millimeter wave detection faces raw data defects. Based on this, the present invention proposes an optional embodiment.

[0078] Step S101 also includes the following specific implementation methods.

[0079] Step S1011: Control the millimeter-wave radar to scan the target area to obtain an original point cloud sequence.

[0080] In an embodiment of the present invention, a millimeter-wave radar is controlled to receive reflected signals from a target area, perform a fast Fourier transform on the original echo, calculate three-dimensional data of range, azimuth, and Doppler, and generate an initial point cloud sequence containing noise and interference.

[0081] Step S1012: performing multi-dimensional dynamic enhancement processing on the original point cloud sequence to obtain the point cloud sequence, wherein the multi-dimensional dynamic enhancement processing includes Doppler dimension filtering, spatial domain motion amplification and / or time domain signal focusing.

[0082] In an embodiment of the present invention, the point cloud quality is reconstructed in three steps. Specifically, static clutter and high-speed interference are filtered out through a velocity threshold to retain the characteristic frequency band of human motion; micro-motion signals are enhanced based on a phase amplification algorithm; and energy information of 10 consecutive frames of point clouds is fused through a sliding window coherent accumulation technique.

[0083] Integrate the 3D processing results and output an enhanced point cloud sequence for subsequent multi-objective processing.

[0084] In the embodiment of the present invention, Doppler filtering can eliminate environmental interference and solve the false positive problem caused by curtain shaking, fan rotation, etc.; spatial motion amplification can break through the physical limit of traditional millimeter-wave radar for micro-motion detection; time domain signal focusing overcomes the limitation of low sampling rate.

[0085] Traditional millimeter wave systems have a spatial conflict absolute failure zone. Based on this, the present invention proposes an optional embodiment.

[0086] The following specific implementation is also included before step S104.

[0087] Step S201 : When a spatially overlapping independent point cloud sequence group appears in a plurality of the independent point cloud sequences, a multi-target trajectory decoupling process is performed on the independent point cloud sequence group.

[0088] In this embodiment of the present invention, the spatial positional relationship of all independent point cloud sequences is monitored in real time. When the minimum bounding boxes of the trajectories of two or more targets overlap by more than 30% in N consecutive frames, it is determined to be a "spatial overlapping sequence group" and the decoupling process is triggered.

[0089] A three-level decoupling is performed on the overlapping sequence groups. Specifically, the motion characteristics of each target in the Doppler dimension are analyzed; an optimization model is established based on the frequency domain sparsity of the mixed signal; and the blind source separation algorithm is applied to reconstruct the independent signal streams.

[0090] The decoupled point cloud sequence is reinjected into the original independent sequence stream and the target motion trajectory database is updated to ensure the continuity of subsequent vital sign detection.

[0091] In the embodiment of the present invention, dynamic trajectory decoupling is used to maintain the accuracy of respiration and heart rate detection in signal overlapping scenarios.

[0092] Traditional millimeter-wave life detection systems face serious limitations in scenarios where multiple targets overlap. When the distance between targets is smaller than the radar resolution, clustering algorithms like DBSCAN cannot effectively separate the target point clouds. Physiological signals in mixed point clouds also generate crosstalk due to overlapping frequency bands, causing existing FFT filtering schemes to mistakenly count other heartbeat signals as belonging to the current target. To address this issue, the present invention proposes an alternative embodiment.

[0093] Step S201 also includes the following specific implementation methods.

[0094] Step S2011 , extracting differentiated motion patterns from the mixed signal corresponding to the independent point cloud sequence group.

[0095] In this embodiment of the present invention, when spatial overlap is detected in multiple independent point cloud sequences, the decoupling process is triggered. First, the joint signal of the multi-target coupling region is extracted from the mixed point cloud data. Time-frequency analysis is then performed to identify the differentiated motion pattern characteristics corresponding to the different targets in the superimposed signal.

[0096] Step S2012: establishing a motion separation model according to the differentiated motion patterns.

[0097] In an embodiment of the present invention, a mathematical model for multi-target motion separation is established based on the extracted differentiated motion pattern features. Specifically, the motion pattern of each target is mapped into an independent signal subspace; the subspace orthogonality constraints are calculated through an optimization algorithm; and a signal reconstruction function with adaptive weights is designed so that the model can dynamically adapt to different motion intensity combinations.

[0098] Step S2013: Process the independent point cloud sequence group according to the motion separation model to obtain a decoupled independent point cloud sequence output by the motion separation model.

[0099] The mixed point cloud sequence group is input into the motion separation model, which performs multi-channel blind source separation on the original signal to eliminate motion interference between targets; the model reallocates the spatial ownership of the point cloud based on the reconstruction results; and the model outputs a decoupled independent point cloud sequence.

[0100] In an embodiment of the present invention, signal disentanglement is achieved by differentiating physiological motion characteristics to address point cloud mixing caused by target spatial intersection. Compared with traditional spatial clustering-based methods, the false detection rate of vital signs in overlapping scenes can be reduced; heart rate detection accuracy can still be maintained in a strong coupling state; and online calculation time can be reduced by presetting a model parameter library.

[0101] like Figure 5FIG2 is a schematic diagram of a terminal device according to an embodiment of the present invention. The terminal device 500 may include a processor 501, a memory 502, and a computer program 503 stored in the memory 502 and executable on the processor 501, such as a multi-target vital sign detection program. When the processor 501 executes the computer program 503, the steps described in the aforementioned multi-target vital sign detection embodiments are implemented.

[0102] The computer program can be divided into one or more modules / units, which are stored in the memory 502 and executed by the processor 501 to implement the present invention. One or more modules / units can be a series of computer program instruction segments that can perform specific functions. These instruction segments are used to describe the execution process of the computer program in the terminal device.

[0103] The terminal device may include, but is not limited to, a processor 501 and a memory 502. Those skilled in the art will appreciate that Figure 5 It is only an example of a terminal device and does not constitute a limitation of the terminal device. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include input and output devices, network access devices, buses, etc.

[0104] The processor 501 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0105] Memory 502 can be an internal storage unit of the terminal device, such as the terminal device's hard drive or memory. Memory 502 can also be an external storage device of the terminal device, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. Furthermore, memory 502 can include both the terminal device's internal storage unit and an external storage device. Memory 502 is used to store computer programs and other programs and data required by the terminal device. Memory 502 can also be used to temporarily store data that has been output or is about to be output.

[0106] It should be noted that, for the convenience and brevity of description, the structure of the above-mentioned terminal device can also refer to the specific description of the structure in the method embodiment, which will not be repeated here.

[0107] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the steps in the multi-target vital sign detection method can be implemented.

[0108] An embodiment of the present invention provides a computer program product. When the computer program product is run on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned multi-target vital sign detection method when executing the computer program product.

[0109] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0110] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0111] In the embodiments provided herein, it should be understood that the disclosed terminal devices and methods can be implemented in other ways. For example, the terminal device embodiments described above are merely illustrative. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via some interface, device, or unit, which may be electrical, mechanical, or other means.

[0112] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0113] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0114] If the integrated module / unit is implemented as a software functional unit and sold or used as a standalone product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the process steps in the above-mentioned method embodiments by using a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signals, telecommunication signals, and software distribution media. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, based on legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0115] The above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may be modified or some of the technical features thereof may be replaced with equivalents. Such modifications or replacements do not deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention and are therefore intended to be included within the scope of protection of the present invention.

Claims

1. A multi-target vital sign detection method, characterized in that: include: Control the millimeter-wave radar to scan the target area and obtain a point cloud sequence; determining the number of target persons in the target area according to the point cloud sequence; When there are multiple target persons, generating an independent point cloud sequence for each target person according to the point cloud sequence; According to the independent point cloud sequence of each target person, the vital sign parameters corresponding to each target person are generated.

2. The multi-target vital sign detection method according to claim 1, wherein: The step of generating the vital sign parameters corresponding to each target person according to the independent point cloud sequence of each target person comprises: extracting point cloud position change information corresponding to tiny physiological movements from the independent point cloud sequence of each target person; Processing the position change information of each point cloud to separate time-varying signals representing different vital signs, wherein the time-varying signals include respiratory motion signals and cardiac motion signals; Each of the time-varying signals is analyzed to obtain vital sign parameters corresponding to each of the target persons, wherein the vital sign parameters include respiratory rate and heart rate parameters.

3. The multi-target vital sign detection method according to claim 2, characterized in that: Before the step of processing the position change information of each point cloud to separate the time-varying signals representing different vital signs, the method further includes: Dynamic interference filtering is performed on each point cloud position change information, where the dynamic interference filtering includes identifying and suppressing noise signals generated by non-physiological movements of the target person.

4. The multi-target vital sign detection method according to claim 1, wherein: The step of determining the number of target persons in the target area according to the point cloud sequence includes: Analyzing the distribution characteristics of the point cloud sequence in the spatial domain; identifying point cloud clusters representing different potential persons based on the distribution characteristics; The number of target point cloud clusters that persist in a continuous time window and meet the characteristics of the target person is counted to obtain the number of the target persons in the target area.

5. The multi-target vital sign detection method according to claim 1, wherein: When there are multiple target persons, the step of generating an independent point cloud sequence for each target person according to the point cloud sequence includes: Based on the point cloud sequence, identifying and tracking spatially separated features of multiple targets; constructing an independent motion trajectory for each target person according to the spatial separation feature; Based on the motion trajectory, the point cloud data belonging to different target persons in the point cloud sequence are separated to obtain an independent point cloud sequence corresponding to each target person.

6. The multi-target vital sign detection method according to claim 1, wherein: The step of controlling the millimeter wave radar to scan the target area to obtain a point cloud sequence includes: Control the millimeter-wave radar to scan the target area and obtain the original point cloud sequence; The original point cloud sequence is subjected to multi-dimensional dynamic enhancement processing to obtain the point cloud sequence, wherein the multi-dimensional dynamic enhancement processing includes Doppler dimension filtering, spatial domain motion amplification and / or time domain signal focusing.

7. The multi-target vital sign detection method according to claim 1, wherein: Before the step of generating the vital sign parameters corresponding to each target person according to the independent point cloud sequence of each target person, the method further includes: When a spatially overlapping independent point cloud sequence group appears in the plurality of independent point cloud sequences, a multi-target trajectory decoupling process is performed on the independent point cloud sequence group.

8. The multi-target vital sign detection method according to claim 7, characterized in that: The step of performing multi-target trajectory decoupling processing on the independent point cloud sequence group includes: extracting differentiated motion patterns from mixed signals corresponding to the independent point cloud sequence groups; establishing a motion separation model based on the differentiated motion patterns; The independent point cloud sequence group is processed according to the motion separation model to obtain a decoupled independent point cloud sequence output by the motion separation model.

9. A terminal device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the multi-target vital sign detection method according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the multi-target vital sign detection method according to any one of claims 1 to 8 are implemented.

Images