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

Abstract

The application discloses a living body detection method and device, a storage medium and an electronic device. The living body detection method comprises tracking each target in a to-be-detected environment by using a millimeter wave radar, and generating trajectory information of each target; at each detection time point of the millimeter wave radar, a PIR sensor is used to detect the living body of the to-be-detected environment synchronously, and a living body detection result is generated; according to the detection time point, the trajectory information is fused with the corresponding living body detection result to generate a perception information time sequence of the target; a cross-validation interval of the corresponding target is obtained according to the perception information time sequence, and self-consistency verification is performed on each cross-validation interval; when all the cross-validation intervals of the target are self-consistent intervals, the target is determined as a living body. The application can improve the accuracy of living body detection.

Description

Technical Field

[0001] This application relates to the field of liveness detection technology, specifically to a liveness detection method, apparatus, storage medium, and electronic device. Background Technology

[0002] With the continuous development of smart home technology, sensor-based target detection systems have been widely used in fields such as home security, environmental monitoring, and automation control. Liveness detection, as a core technology in smart home systems, primarily aims to identify and distinguish human bodies from other non-living targets in order to achieve accurate environmental monitoring and intelligent response.

[0003] Traditional liveness detection methods often rely on a single sensor. However, a single sensor often struggles to provide reliable detection in the face of various interference factors and complex dynamic scenarios. This is particularly true for detecting stationary or slightly moving targets, where false positives are common, resulting in low accuracy. Summary of the Invention

[0004] This application provides a liveness detection method, apparatus, storage medium, and electronic device, which can improve the accuracy of liveness detection.

[0005] In a first aspect, embodiments of this application provide a liveness detection method, including:

[0006] Millimeter-wave radar is used to track each target in the environment to be detected, and trajectory information of each target is generated;

[0007] At each detection time point of the millimeter-wave radar, a PIR sensor is used simultaneously to perform liveness detection on the environment to be detected, and a liveness detection result is generated.

[0008] Based on the detection time point, the trajectory information is fused with the corresponding liveness detection result to generate a time series of perception information of the target;

[0009] Based on the time series of the perceived information, obtain the cross-validation intervals corresponding to the target, and perform self-consistency verification on each cross-validation interval;

[0010] The target is determined to be a live object when all of the cross-validation intervals of the target are self-consistent intervals.

[0011] In the liveness detection method provided in this application embodiment, the trajectory information includes several trajectory elements. The step of fusing the trajectory information with the corresponding liveness detection result based on the detection time point to generate a time series of perception information for the target includes:

[0012] Based on the detection time point, the liveness detection result at the same moment is paired with the trajectory element to form a sensing element;

[0013] Arrange the aforementioned sensing elements in chronological order to generate a time sequence of sensing information for the target.

[0014] In the liveness detection method provided in this application embodiment, the step of obtaining the cross-validation interval corresponding to the target based on the time series of the perceived information, and performing self-consistency verification on each cross-validation interval, includes:

[0015] Filter out continuous time periods that meet preset motion criteria from the time series of the perceived information as cross-validation intervals;

[0016] Self-consistency verification is performed on each of the cross-validation intervals.

[0017] In the liveness detection method provided in this application embodiment, the self-consistency verification of each of the cross-validation intervals includes:

[0018] For each of the aforementioned cross-validation intervals, the liveness detection results contained therein are verified;

[0019] When at least one of the liveness detection results in a cross-validation interval is characterized as a liveness, the cross-validation interval is determined to be a self-consistent interval.

[0020] When all the liveness detection results in a cross-validation interval are characterized as non-liveness, the cross-validation interval is determined to be a non-consistent interval.

[0021] In the liveness detection method provided in this application embodiment, the step of selecting continuous time periods that satisfy preset motion criteria from the time series of the perceived information as cross-validation intervals includes:

[0022] Based on preset displacement criteria and / or velocity criteria, continuous time periods that meet the conditions are selected from the time series of the sensed information.

[0023] The consecutive time periods that meet the conditions are used as cross-validation intervals.

[0024] In the liveness detection method provided in this application embodiment, the step of filtering out continuous time periods that meet the conditions from the time series of the sensed information based on preset displacement criteria and / or velocity criteria includes:

[0025] The time series of the sensed information is traversed based on a preset speed threshold and / or a preset displacement threshold.

[0026] When there is a consecutive time period in the time series of the sensed information where the absolute value of the instantaneous velocity of multiple sensed elements exceeds a preset velocity threshold, the corresponding consecutive time period is determined to meet the condition; and / or

[0027] When there is a continuous time period in the time series of the sensed information where the displacement change exceeds a preset displacement threshold, the corresponding continuous time period is determined to meet the condition.

[0028] In the liveness detection method provided in this application embodiment, the step of tracking each target in the detection environment using millimeter-wave radar and generating trajectory information for each target includes:

[0029] It receives and processes the echo signals from millimeter-wave radar to obtain point cloud data of one or more targets in the environment to be detected.

[0030] The point cloud data is clustered to distinguish different targets;

[0031] A multi-target tracking algorithm is used to continuously track each of the clustered targets and generate trajectory information for each target.

[0032] Secondly, embodiments of this application provide a liveness detection device, comprising:

[0033] The tracking unit is used to track each target in the environment to be detected using millimeter-wave radar and generate trajectory information for each target.

[0034] The detection unit is used to synchronously perform liveness detection on the environment to be detected using a PIR sensor at each detection time point of the millimeter-wave radar, and generate liveness detection results.

[0035] The fusion unit is used to fuse the trajectory information with the corresponding liveness detection result according to the detection time point to generate a time series of perception information of the target;

[0036] The verification unit is used to obtain the cross-validation interval corresponding to the target based on the time series of the perceived information, and to perform self-consistency verification on each cross-validation interval.

[0037] A determining unit is configured to determine that the target is a living entity when all of the cross-validation intervals of the target are self-consistent intervals.

[0038] Thirdly, this application provides a storage medium storing a plurality of instructions adapted for loading by a processor to execute any of the above-described liveness detection methods.

[0039] Fourthly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the liveness detection method described in any of the above claims.

[0040] In summary, the liveness detection method provided in this application includes: tracking targets in the environment to be detected using millimeter-wave radar to generate trajectory information for each target; simultaneously performing liveness detection on the environment to be detected using a PIR sensor at each detection time point of the millimeter-wave radar to generate a liveness detection result; fusing the trajectory information with the corresponding liveness detection result according to the detection time point to generate a time series of perception information for the target; obtaining cross-validation intervals corresponding to the target based on the time series of perception information, and performing self-consistency verification on each cross-validation interval; determining that the target is a live target when all cross-validation intervals of the target are self-consistent intervals. This application, through synchronous detection and data fusion of millimeter-wave radar and PIR sensors, and by introducing a time series-based "cross-validation interval" decision mechanism, fundamentally improves the accuracy of liveness detection. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram illustrating an application scenario of the liveness detection method provided in the embodiments of this application.

[0043] Figure 2 This is a schematic flowchart of the liveness detection method provided in the embodiments of this application.

[0044] Figure 3 This is a schematic diagram of the liveness detection device provided in the embodiments of this application.

[0045] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0046] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0047] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, components, features, and elements with the same names in different embodiments of this application may have the same meaning or different meanings, the specific meaning of which must be determined by its interpretation in that specific embodiment or further in conjunction with the context of that specific embodiment.

[0048] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0049] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.

[0050] In the description of this application, it should be noted that the terms "upper," "lower," "left," "right," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. In addition, terms such as "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0051] Traditional liveness detection methods often rely on a single sensor. However, a single sensor often struggles to provide reliable detection in the face of various interference factors and complex dynamic scenarios. This is particularly true for detecting stationary or slightly moving targets, where false positives are common, resulting in low accuracy.

[0052] Based on this, embodiments of this application provide a liveness detection method, apparatus, storage medium, and electronic device. Specifically, the liveness detection apparatus can be integrated into an electronic device, which can be a server or a terminal, etc. The terminal can include mobile phones, wearable smart devices, tablet computers, laptops, and personal computers (PCs), etc. The server can be a single server or a server cluster composed of multiple servers, and can be a physical server or a virtual server.

[0053] For example, such as Figure 1 As shown, the electronic device can use millimeter-wave radar to track each target in the environment to be detected and generate trajectory information for each target; at each detection time point of the millimeter-wave radar, a PIR sensor is used simultaneously to perform liveness detection in the environment to be detected and generate liveness detection results; based on the detection time point, the trajectory information and the corresponding liveness detection results are fused to generate a time series of target perception information; based on the time series of perception information, the cross-validation intervals of the corresponding targets are obtained, and self-consistency verification is performed on each cross-validation interval; when all cross-validation intervals of the target are self-consistent intervals, the target is determined to be a live object.

[0054] The technical solutions shown in this application will be described in detail below through specific embodiments. It should be noted that the order of description of the following embodiments is not intended to limit the priority of the embodiments.

[0055] Please see Figure 2 , Figure 2 This is a schematic flowchart of the liveness detection method provided in this application embodiment. The specific flow of the liveness detection method can be as follows:

[0056] 101. Use millimeter-wave radar to track each target in the environment to be detected and generate trajectory information for each target.

[0057] Millimeter-wave radar can transmit frequency-modulated continuous wave (FMCW) signals and receive their echo signals, thereby enabling active perception of the environment to be detected and generating trajectory information for each target.

[0058] Specifically, it can receive and process the echo signals from millimeter-wave radar to obtain point cloud data of one or more targets in the environment to be detected; perform clustering processing on the point cloud data to distinguish different targets; and continuously track each clustered target through a multi-target tracking algorithm to generate trajectory information for each target.

[0059] In some embodiments, the echo signal received by the millimeter-wave radar can be processed by mixing, amplification, and analog-to-digital conversion. Then, by Fast Fourier Transform (FFT) processing, the distance and azimuth information of each scattering point in the environment to be detected relative to the millimeter-wave radar can be extracted. The radial velocity can be directly obtained by Doppler frequency shift analysis, and finally a frame of point cloud data containing distance, angle, and velocity information is formed.

[0060] Since a single target is usually composed of multiple scattering points, clustering algorithms (such as DBSCAN and K-Means) can be used to process single-frame point cloud data, aggregating scattering points belonging to the same physical target together to form independent target clusters, thereby distinguishing different targets.

[0061] To achieve continuous tracking of multiple targets, multi-target tracking algorithms (such as multi-hypothesis tracking (MHT) based on Kalman filtering or joint probabilistic data association (JPDA)) can be used to track each target.

[0062] In some embodiments, when a new target is generated in a new frame, a multi-target tracking algorithm can be used to correlate these new targets with targets already existing in the previous frame (e.g., using a nearest neighbor algorithm) to determine if they are the same target. For successfully correlated targets, a Kalman filter can be used to predict their position coordinates and velocity information at the next moment, and then updated and corrected using the actual measurements of the current frame, thereby outputting a smooth and continuous trajectory. Furthermore, a unique identifier (ID) can be assigned to each newly appearing target, and it can be tracked throughout its entire lifecycle, from its appearance, residence or movement in the environment, until it disappears.

[0063] Finally, a trajectory information can be output for each target ID. This trajectory information is a time series and includes several trajectory elements, each of which includes, but is not limited to, the detection time point, location coordinates, and instantaneous velocity.

[0064] In some embodiments, the position coordinates can be the target's coordinates (x, y) in a Cartesian coordinate system. The instantaneous velocity is the target's velocity vector at the detection time point, which can be further decomposed into two key components: radial velocity and tangential velocity.

[0065] The radial velocity can be directly measured by the millimeter-wave radar through the Doppler effect, representing the velocity component of the target approaching or moving away from the millimeter-wave radar. The tangential velocity can be calculated by comparing the changes in position coordinates (i.e., displacement) between consecutive frames and combining them with the time difference, representing the lateral velocity component of the target perpendicular to the radial direction of the millimeter-wave radar.

[0066] 102. At each detection time point of the millimeter-wave radar, the PIR sensor is used simultaneously to perform liveness detection on the environment to be detected and generate liveness detection results.

[0067] In some embodiments, during each environmental detection, the main control chip of the millimeter-wave radar generates a high-level pulse signal via a general-purpose input / output (GPIO) pin as a synchronization trigger signal. This synchronization trigger signal is directly transmitted to the signal processing circuit of the passive infrared (PIR) sensor. The PIR sensor is configured to operate in trigger mode, and upon receiving this synchronization trigger signal, it immediately initiates a liveness detection of the environment to be detected.

[0068] By using this hardware synchronization method of "millimeter-wave radar as the main driver and PIR sensor as the secondary driver", it can be ensured that every detection time point of the millimeter-wave radar has a liveness detection result that is perfectly aligned in time, and the time error between the two can be controlled at the millisecond or even microsecond level.

[0069] In another embodiment, during power-on initialization, the main control microprocessor (MCU) can simultaneously power on the millimeter-wave radar and the PIR sensor, and initialize their operating programs, enabling both to begin operation. The main control MCU can internally set a fixed sampling period T (e.g., T = 100 milliseconds, i.e., 10 samples per second). This sampling period serves as a global synchronization clock reference.

[0070] At the beginning of each sampling cycle, the main control MCU can read the data collected by the millimeter-wave radar and PIR sensor in the current cycle in parallel or sequentially at extremely fast speeds via software instructions. Since the reading actions are controlled by the same clock source and the time interval is extremely short, it can be assumed that the two sets of data were collected within the same time window, thus achieving synchronization at the software level.

[0071] After reading two sets of data, the main control MCU can assign the same timestamp (detection time point) to both sets. This timestamp is typically the start time of the current sampling period or the time when data reading is completed. By aligning the timestamps, even if there is a slight phase difference between the sampling clocks inside the millimeter-wave radar and the PIR sensor, they can be considered as data from the same detection time point at the system level.

[0072] 103. Based on the detection time point, the trajectory information is fused with the corresponding liveness detection results to generate a time series of target perception information.

[0073] In some embodiments, based on the detection time point, the liveness detection results at the same moment can be paired with trajectory elements to form sensing elements; then, several sensing elements are arranged in chronological order to generate a time sequence of sensing information of the target.

[0074] For example, for each target tracked by millimeter-wave radar, the trajectory element generated at a certain detection time point can be obtained first. This trajectory element contains the target's position coordinates and instantaneous velocity at that detection time point. Simultaneously, the liveness detection result corresponding to that detection time point is retrieved from the PIR sensor's output buffer. Combining the trajectory element and the liveness detection result forms a sensing element.

[0075] For each target, whenever a new perceptual element is generated, it can be arranged into the perceptual information time series of the corresponding target in chronological order.

[0076] 104. Obtain the cross-validation intervals of the corresponding targets based on the time series of perceived information, and perform self-consistency verification on each cross-validation interval.

[0077] Specifically, continuous time periods that meet preset motion criteria can be selected from the time series of perceived information as cross-validation intervals; then, self-consistency verification can be performed on each cross-validation interval.

[0078] The preset motion judgment may include velocity criteria and / or displacement criteria. Therefore, the step "selecting continuous time periods that meet the preset motion criteria from the time series of perceived information as cross-validation intervals" can specifically be: selecting continuous time periods that meet the conditions from the time series of perceived information based on preset displacement criteria and / or velocity criteria; and using the continuous time periods that meet the conditions as cross-validation intervals.

[0079] In some embodiments, a preset speed threshold (e.g., 0.15 m / s) can be set. When the absolute value of the instantaneous speed of N consecutive (N≥1, usually set to 3-5 according to the sampling rate) sensing elements in a continuous time period of the sensing information time series exceeds the preset speed threshold, the continuous time period is identified as a cross-validation interval.

[0080] In some embodiments, a preset displacement threshold (e.g., 0.5 meters) can be set. When the displacement change exceeds the preset displacement threshold within a continuous time period of the sensing information time series, the continuous time period is identified as a cross-validation interval. Here, the displacement change refers to the displacement value between the starting position and the ending position of the target within the continuous time period (i.e., the straight-line distance between the position coordinates of the starting sensing element and the position coordinates of the ending sensing element within the continuous time period).

[0081] In some embodiments, only the tangential displacement (i.e., the displacement component perpendicular to the radial direction of the millimeter-wave radar) can be used as the displacement change, because this is a typical characteristic of behaviors such as lateral walking of the human body, and can be effectively distinguished from many non-living targets that mainly move radially.

[0082] That is, the step "selecting continuous time periods that meet the conditions from the sensing information time series based on preset displacement criteria and / or velocity criteria" can be: traversing the sensing information time series based on preset velocity thresholds and / or preset displacement thresholds; when there are continuous time periods in the sensing information time series where the absolute value of the instantaneous velocity of multiple consecutive sensing elements exceeds the preset velocity threshold, determining that the corresponding continuous time periods meet the conditions; and / or when there are continuous time periods in the sensing information time series where the displacement change exceeds the preset displacement threshold, determining that the corresponding continuous time periods meet the conditions.

[0083] In this context, performing self-consistency verification on each cross-validation interval refers to determining whether there are any liveness detection results within that cross-validation interval that characterize a liveness.

[0084] If at least one liveness detection result indicates a liveness, then the cross-validation interval is determined to be a self-consistent interval. This indicates that when the target undergoes significant movement, the PIR sensor also confirms its liveness attribute, demonstrating consistency between the millimeter-wave radar and the PIR sensor's perception.

[0085] If all liveness detection results within a cross-validation interval are characterized as non-liveness, then the cross-validation interval is determined to be a non-consistent interval. This indicates that a target undergoing significant motion was not identified as a live object by the PIR sensor, exhibiting the characteristic of "moving but not generating heat," and is highly likely to be a non-live interference object.

[0086] That is, the step "perform self-consistency verification for each cross-validation interval" can be: for each cross-validation interval, verify the liveness detection results contained therein; when at least one liveness detection result in a cross-validation interval is characterized as a liveness, the cross-validation interval is determined to be a self-consistent interval; when all liveness detection results in a cross-validation interval are characterized as non-liveness, the cross-validation interval is determined to be a non-self-consistent interval.

[0087] This embodiment introduces the concept of "cross-validation intervals" to deeply correlate originally independent trajectory information with discrete liveness detection within a discriminative time window. This not only determines the instantaneous state of the target but also makes comprehensive judgments by analyzing its historical behavior segments (cross-validation intervals). It can effectively handle complex multi-target scenarios (such as the simultaneous presence of a person and a robot vacuum cleaner) and maintain an extremely low false alarm rate.

[0088] 105. When all cross-validation intervals of the target are self-consistent intervals, the target is determined to be a live organism.

[0089] Through the above embodiments, each tracked target can be labeled with a liveness status identifier. When the final identity of the target needs to be output (e.g., reported to a smart home system), the self-consistency verification results of all historical cross-validation intervals that have been identified and verified throughout the entire tracking period can be checked. The target is determined to be live only if all historical cross-validation intervals of the target are self-consistent intervals. That is, every time the target exhibits significant movement in history (thus triggering cross-validation), its behavior is confirmed by the PIR sensor (at least one liveness detection result characterizes it as live). This high degree of behavioral consistency is a typical characteristic of live targets (such as humans).

[0090] In this embodiment, once a target is identified as a living entity, its living state is continuously maintained. Even if the target subsequently enters a static state (no longer generating new cross-validation intervals), it will still be reported as a living entity until the target disappears or its state reverses. This embodiment ensures the continuity of human presence perception. Users will not experience frequent switching of the living state due to brief periods of stillness (such as sitting), thus achieving smooth and natural smart home control.

[0091] State reversal refers to the occurrence of a new, verified, inconsistent cross-validation interval during subsequent tracking of the target. In this case, its liveness state should be immediately reversed to "non-liveness." Target disappearance refers to the target leaving the detection environment and being determined as disappeared by the millimeter-wave radar, and its target ID is removed.

[0092] In this embodiment, if a target is determined to be inconsistent in any historical cross-validation interval, it will be immediately and permanently classified as a non-living target. This state is irreversible throughout its entire tracking cycle. This embodiment adopts the strict standard that "all cross-validation intervals must be consistent" and grants a veto power of "one instance of inconsistency results in permanent classification as non-living," which greatly reduces the probability of misclassifying a non-living target as a living target from an algorithmic perspective. This effectively solves the misclassification problem in multi-target scenarios (such as a person and a robot vacuum cleaner running in parallel). Even if a non-living target happens to be close to the trajectory of a living target, its own historical "inconsistency" records can ensure that it is correctly distinguished.

[0093] In summary, the liveness detection method provided in this application includes: tracking targets in the detection environment using millimeter-wave radar to generate trajectory information for each target; simultaneously performing liveness detection in the detection environment using a PIR sensor at each detection time point of the millimeter-wave radar to generate liveness detection results; fusing the trajectory information with the corresponding liveness detection results according to the detection time points to generate a target perception information time series; obtaining the cross-validation intervals of the corresponding targets based on the perception information time series, and performing self-consistency verification on each cross-validation interval; determining that the target is a live target when all cross-validation intervals of the target are self-consistent intervals. This application, through synchronous detection and data fusion of millimeter-wave radar and PIR sensors, and by introducing a time series-based "cross-validation interval" decision mechanism, fundamentally improves the accuracy of liveness detection.

[0094] To facilitate better implementation of the liveness detection method provided in this application, this application also provides a liveness detection device. The meanings of the terms used are the same as in the liveness detection method described above, and specific implementation details can be found in the descriptions within the method embodiments.

[0095] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of the liveness detection device provided in an embodiment of this application. The liveness detection device may include a tracking unit 201, a detection unit 202, a fusion unit 203, a verification unit 204, and a determination unit 205.

[0096] The tracking unit 201 is used to track each target in the environment to be detected using millimeter-wave radar and generate trajectory information for each target.

[0097] The detection unit 202 is used to synchronously perform liveness detection on the environment to be detected using a PIR sensor at each detection time point of the millimeter-wave radar and generate liveness detection results.

[0098] The fusion unit 203 is used to fuse trajectory information with the corresponding liveness detection results based on the detection time point to generate a time series of target perception information.

[0099] The verification unit 204 is used to obtain the cross-validation interval of the corresponding target based on the time series of the perceived information, and to perform self-consistency verification on each cross-validation interval.

[0100] The determination unit 205 is used to determine that the target is a live target when all cross-validation intervals of the target are self-consistent intervals.

[0101] For specific implementation methods of each of the above units, please refer to the embodiments of the above-mentioned liveness detection method, which will not be repeated here.

[0102] In summary, the liveness detection device provided in this application embodiment can use the tracking unit 201 to track each target in the detection environment using millimeter-wave radar, generating trajectory information for each target; the detection unit 202 simultaneously uses a PIR sensor to perform liveness detection in the detection environment at each detection time point of the millimeter-wave radar, generating liveness detection results; the fusion unit 203 fuses the trajectory information with the corresponding liveness detection results according to the detection time points, generating a time series of target perception information; the verification unit 204 obtains the cross-validation intervals of the corresponding target based on the time series of perception information, and performs self-consistency verification on each cross-validation interval; the determination unit 205 determines that the target is a liveness target when all cross-validation intervals of the target are self-consistent intervals. This application embodiment fundamentally improves the accuracy of liveness detection by synchronous detection and data fusion of millimeter-wave radar and PIR sensors, and by introducing a time series-based "cross-validation interval" decision mechanism.

[0103] This application also provides an electronic device that may integrate the liveness detection device described in this application, such as... Figure 4 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically:

[0104] The electronic device may include components such as a processor 301 with one or more processing cores and a memory 302 with one or more computer-readable storage media. Those skilled in the art will understand that... Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein:

[0105] The processor 301 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs stored in the memory 302 and / or this application, and by calling data stored in the memory 302, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Optionally, the processor 301 may include one or more processing cores; preferably, the processor 301 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operation of the storage medium, user interface, and application programs, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 301.

[0106] The memory 302 can be used to store software programs and this application. The processor 301 executes various functional applications and data processing by running the software programs and this application stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store applications required for operating the storage medium and at least one function; the data storage area may store data created based on the use of the electronic device. In addition, the memory 302 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 302 may also include a memory controller to provide the processor 301 with access to the memory 302.

[0107] Although not shown, the electronic device may also include a display unit, an input unit, and a power supply, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 301 in the electronic device loads the executable files corresponding to the processes of one or more application programs into the memory 302 according to the following instructions, and the processor 301 runs the application programs stored in the memory 302 to realize various functions, as follows:

[0108] Millimeter-wave radar is used to track various targets in the environment to be detected and generate trajectory information for each target.

[0109] At each detection time point of the millimeter-wave radar, a PIR sensor is used simultaneously to perform liveness detection on the environment to be detected and generate liveness detection results.

[0110] Based on the detection time point, the trajectory information is fused with the corresponding liveness detection results to generate a time series of target perception information;

[0111] The cross-validation intervals for the corresponding targets are obtained based on the time series of perceived information, and the self-consistency of each cross-validation interval is verified.

[0112] The target is determined to be a live target when all cross-validation intervals of the target are self-consistent intervals.

[0113] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0114] Therefore, embodiments of this application provide a storage medium storing a plurality of instructions that can be loaded by a processor to execute steps in any of the methods provided in embodiments of this application. For example, the instructions can execute the following steps:

[0115] Millimeter-wave radar is used to track various targets in the environment to be detected and generate trajectory information for each target.

[0116] At each detection time point of the millimeter-wave radar, a PIR sensor is used simultaneously to perform liveness detection on the environment to be detected and generate liveness detection results.

[0117] Based on the detection time point, the trajectory information is fused with the corresponding liveness detection results to generate a time series of target perception information;

[0118] The cross-validation intervals for the corresponding targets are obtained based on the time series of perceived information, and the self-consistency of each cross-validation interval is verified.

[0119] The target is determined to be a live target when all cross-validation intervals of the target are self-consistent intervals.

[0120] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0121] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0122] Since the instructions stored in the storage medium can execute the steps of any method provided in the embodiments of this application, the beneficial effects that any method provided in the embodiments of this application can achieve can be realized. For details, please refer to the previous embodiments, which will not be repeated here.

[0123] The above provides a detailed description of the liveness detection method, apparatus, storage medium, and electronic device provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of this application. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for detecting liveness, characterized in that, include: Millimeter-wave radar is used to track each target in the environment to be detected, and trajectory information of each target is generated, the trajectory information including several trajectory elements; At each detection time point of the millimeter-wave radar, a PIR sensor is used simultaneously to perform liveness detection on the environment to be detected, and a liveness detection result is generated. Based on the detection time point, the liveness detection result at the same moment is paired with the trajectory element to form a sensing element; Arrange the aforementioned sensing elements in chronological order to generate a time sequence of sensing information for the target. Filter out continuous time periods that meet preset motion criteria from the time series of the perceived information as cross-validation intervals; For each of the aforementioned cross-validation intervals, the liveness detection results contained therein are verified; When at least one of the liveness detection results in a cross-validation interval is characterized as a liveness, the cross-validation interval is determined to be a self-consistent interval. When all the liveness detection results in a cross-validation interval are characterized as non-liveness, the cross-validation interval is determined to be a non-consistent interval. The target is determined to be a live object when all of the cross-validation intervals of the target are self-consistent intervals.

2. The live detection method as described in claim 1, characterized in that, The step of selecting continuous time periods that satisfy preset motion criteria from the time series of the perceived information as cross-validation intervals includes: Based on preset displacement criteria and / or velocity criteria, continuous time periods that meet the conditions are selected from the time series of the sensed information. The consecutive time periods that meet the conditions are used as cross-validation intervals.

3. The live detection method as described in claim 2, characterized in that, The step of filtering continuous time periods that meet the conditions from the time series of the sensed information based on preset displacement and / or velocity criteria includes: The time series of the sensed information is traversed based on a preset speed threshold and / or a preset displacement threshold. When there is a consecutive time period in the time series of the sensed information where the absolute value of the instantaneous velocity of multiple sensed elements exceeds a preset velocity threshold, the corresponding consecutive time period is determined to meet the condition; and / or When there is a continuous time period in the time series of the sensed information where the displacement change exceeds a preset displacement threshold, the corresponding continuous time period is determined to meet the condition.

4. The live detection method as described in claim 1, characterized in that, The process of using millimeter-wave radar to track targets in the environment to be detected and generating trajectory information for each target includes: It receives and processes the echo signals from millimeter-wave radar to obtain point cloud data of one or more targets in the environment to be detected. The point cloud data is clustered to distinguish different targets; A multi-target tracking algorithm is used to continuously track each of the clustered targets and generate trajectory information for each target.

5. A liveness detection device, characterized in that, include: The tracking unit is used to track each target in the environment to be detected using millimeter-wave radar and generate trajectory information for each target, the trajectory information including several trajectory elements; The detection unit is used to synchronously perform liveness detection on the environment to be detected using a PIR sensor at each detection time point of the millimeter-wave radar, and generate liveness detection results. The fusion unit is used to pair the liveness detection results at the same time with the trajectory elements based on the detection time point to form a sensing element; and to arrange a number of the sensing elements in chronological order to generate a time sequence of the sensing information of the target. The verification unit is used to select continuous time periods that meet preset motion criteria from the time series of the perceived information as cross-verification intervals; for each cross-verification interval, it verifies the liveness detection results contained therein; when at least one liveness detection result in a cross-verification interval indicates a liveness, the cross-verification interval is determined to be a self-consistent interval; when all liveness detection results in a cross-verification interval indicate a non-liveness, the cross-verification interval is determined to be a non-self-consistent interval. A determining unit is configured to determine that the target is a living entity when all of the cross-validation intervals of the target are self-consistent intervals.

6. A storage medium, characterized in that, The storage medium stores multiple instructions, which are adapted for loading by a processor to execute the liveness detection method according to any one of claims 1-4.

7. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the liveness detection method as described in any one of claims 1-4.

Images