A state identification method and device

Abstract

The embodiment of the present application provides a state recognition method and device, which relates to the field of data processing technology, and can identify the user's bed status without infringing the user's personal privacy. The state recognition method includes: obtaining multiple frames of first echo signals within a detection period through a radar device; processing the multiple frames of first echo signals within the detection period to obtain processed multiple frames of first echo signals, and the signal-to-noise ratio of the processed first echo signals is greater than the signal-to-noise ratio of the first echo signals before processing; when it is determined that there is a target human body in the target area, according to the processed multiple frames of first echo signals, determine the position information of the target human body; obtain the position information of the bed in the target area, and determine the target human body's bed status according to the position information of the bed, the position information of the target human body, and a pre-state detection model for detecting the target human body's bed status.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a state identification method and device. Background Art

[0002] With the improvement of living standards and changes brought about by consumption upgrades, intelligence, technology, and artificial intelligence are developing rapidly, and human state recognition technology is gradually maturing. At present, human state recognition technology is widely used in many fields, for example, it can be applied in the field of smart homes. In this way, when the smart home recognizes that there is no one in the house, it can execute the shutdown action to achieve the effect of saving energy.

[0003] Existing state recognition methods usually obtain user image information based on a camera, so that the user's state can be recognized based on the image information. However, this method has the potential risk of privacy leakage and may cause information security issues. In addition, the camera is greatly affected by the environment. Environmental factors such as lighting, smoke, and occlusion will reduce the accuracy of the recognition result. Summary of the invention

[0004] The embodiments of the present application provide a state recognition method and device, which can recognize the state of a user without infringing the user's personal privacy.

[0005] In a first aspect, an embodiment of the present application provides a state recognition method, the method comprising: acquiring multiple frames of first echo signals within a detection period through a radar device; processing the multiple frames of first echo signals within the detection period to obtain multiple frames of processed first echo signals, wherein the signal-to-noise ratio of the processed first echo signals is greater than the signal-to-noise ratio of the first echo signals before processing; when it is determined that a target human body exists in a target area, determining position information of the target human body according to the processed multiple frames of first echo signals; acquiring position information of a bed in the target area, and determining the in-bed state of the target human body according to the position information of the bed, the position information of the target human body, and a pre-state detection model for detecting the in-bed state of the target human body.

[0006] The state recognition method provided in the embodiment of the present application can process each frame of echo signal of the target area within a preset period acquired by the radar device to obtain a multi-frame echo signal with a high signal-to-noise ratio. Then, based on the multi-frame echo signal, determine whether there is a target human body in the target area. When it is determined that there is a human body in the target area, the position information of the target human body can be determined based on the processed multi-frame echo signal. Further, after determining the position information of the bed, the in-bed state of the target human body is determined based on the position information of the target human body and the position information of the bed and a preset state detection model. The state recognition method provided in the embodiment of the present application does not require the use of a shooting device, and can monitor the user's behavior and state during sleep in real time while protecting the user's personal privacy. In addition, the radar equipment will not be affected by environmental problems such as light, so this method is more practical and universal.

[0007] In a second aspect, a state recognition device is provided, including: an acquisition unit and a processing unit; the acquisition unit is used to acquire multiple frames of first echo signals within a detection period through a radar device; the processing unit is used to process the multiple frames of first echo signals within the detection period to obtain the processed multiple frames of first echo signals, and the signal-to-noise ratio of the processed first echo signals is greater than the signal-to-noise ratio of the first echo signals before processing; the processing unit is also used to determine the position information of the target human body according to the processed multiple frames of first echo signals when it is determined that there is a target human body in the target area; the acquisition unit is also used to acquire the position information of a bed in the target area; the processing unit is also used to determine the in-bed state of the target human body according to the position information of the bed, the position information of the target human body and a pre-state detection model for detecting the in-bed state of the target human body.

[0008] In a third aspect, a state identification device is provided, comprising: one or more processors and one or more memories; wherein the one or more memories are used to store computer program code, the computer program code includes computer instructions, and one or more processors execute the computer instructions to enable the state identification device to perform any one of the methods provided in the first aspect above.

[0009] According to a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are executed on a computer, the computer executes any one of the methods provided in the first aspect.

[0010] According to a fifth aspect, a computer program product comprising computer instructions is provided. When the computer instructions are executed on a computer, the computer is enabled to execute any one of the methods provided in the first aspect.

[0011] The beneficial effects brought about by the description of the second to fifth aspects above can be referred to the analysis of the beneficial effects of the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A flow chart of a state identification method provided in an embodiment of the present application;

[0013] Figure 2 A flowchart of another state identification method provided in an embodiment of the present application;

[0014] Figure 3 A flowchart of another state identification method provided in an embodiment of the present application;

[0015] Figure 4 A flowchart of a method for processing an echo signal provided in an embodiment of the present application;

[0016] Figure 5 A schematic diagram of an azimuth angle provided in an embodiment of the present application;

[0017] Figure 6 A flowchart of a method for determining position information of a bed provided in an embodiment of the present application;

[0018] Figure 7 A flowchart of a method for determining the location information of a target human body provided in an embodiment of the present application;

[0019] Figure 8 A schematic diagram of a training method for a pre-state detection model provided in an embodiment of the present application;

[0020] Fig. 9 A schematic diagram of the composition of a state recognition device provided in an embodiment of the present application;

[0021] Fig.10 A schematic diagram of the hardware structure of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0023] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0024] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances. In addition, when describing a pipeline, the "connected" and "connection" used in this application have the meaning of conduction. The specific meaning needs to be understood in conjunction with the context.

[0025] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0026] To facilitate understanding, some basic concepts of terms or technologies involved in the embodiments of the present invention are first briefly introduced and explained.

[0027] 1. Radar

[0028] Radar is an electronic device that uses electromagnetic waves to detect targets. The radar's transmitting antenna converts the high-frequency current signal generated by the transmitter circuit or the guided wave on the transmission line into an electromagnetic wave with a certain polarization mode that can be transmitted in space and is transmitted along a preset direction. When the electromagnetic wave encounters an obstacle in the forward direction, part of the electromagnetic wave will be reflected back in the opposite direction of the transmission direction. At this time, the radar's receiving antenna can receive the reflected electromagnetic wave and convert it into a high-frequency current signal or a transmission line guided wave. By subsequently processing the obtained echo signal, the state information such as the distance, speed and angle of the detection target (that is, the target human body in this application) can be extracted.

[0029] Among them, millimeter wave radar is a radar that works in the millimeter wave band, which can emit signals with a wavelength of 1-10mm and a frequency of 30GHZ-300GHZ. In the electromagnetic spectrum, this wavelength is considered a short wavelength, and a short wavelength means high accuracy. Millimeter wave systems with an operating frequency of 76-81GHz (corresponding to a wavelength of about 4mm) will be able to detect movements as small as a few tenths of a millimeter.

[0030] 2. Pulse Compression (Linear Frequency Modulation Continuous Wave, LFMCW)

[0031] Pulse compression refers to the process of processing the echo signal of a wide pulse signal to obtain a narrow pulse, so that the processed signal maintains the high distance resolution of the narrow pulse and obtains the strong detection capability of the wide pulse.

[0032] Radar equipment usually transmits ultra-large time-width and bandwidth signals at the transmitting end to improve the speed measurement accuracy and speed resolution of radar signals. However, ultra-large time-width signals cannot be pulse compressed in the time domain, so the signal needs to be transformed into the frequency domain for processing. Since pulse compression can compress the time domain signal to the frequency for processing, the signal-to-noise ratio of the echo signal is greatly improved. The signal gain obtained in this process is TB. Where T is the sweep time width and B is the sweep bandwidth.

[0033] Based on this, at the receiving end of the radar, the wide pulse signal can be compressed into a narrow pulse signal through LFMCW, which can improve the radar's distance resolution accuracy and distance resolution for the target.

[0034] 3. Moving Target Indication (MTI)

[0035] Moving target display technology is one of the earliest technologies used for clutter suppression. It is based on the difference in the Doppler effect between moving targets and clutter in the spectrum, and uses stopband filtering to suppress the clutter spectrum, thereby extracting the echo of the detected target from the clutter background. It is a kind of radar overall technology.

[0036] In addition to noise interference, there are also a large number of clutter signals in the echo signal. These clutter signals are mainly caused by the environment where the radar is located, such as: ground, walls, tables and chairs and other furniture. The biggest difference between these clutter signals and the target echo signal is Doppler. Therefore, they can be detected and suppressed in the Doppler domain. MTI technology can perform two-pulse cancellation on the received data to eliminate the interference and influence of stationary clutter.

[0037] 4. Accumulation

[0038] Accumulation is a commonly used processing method in signal processing, which is mainly divided into coherent accumulation and incoherent accumulation. Coherent accumulation generally refers to the accumulation of complex signals, and incoherent accumulation generally refers to the accumulation of signal amplitudes. Incoherent accumulation can only retain the modulus value, exclude strict phase relationships, and improve the signal-to-noise ratio of the processed signal.

[0039] 5. Constant False Alarm Rate (CFAR)

[0040] Constant false alarm detection technology is a technology that the radar system uses to distinguish the signal and noise output by the receiver under the condition of keeping the false alarm probability constant to determine whether the target signal exists. The principle is to first process the input noise and then determine a threshold, and compare this threshold with the input signal. If the input signal exceeds this threshold, it is judged as a target, otherwise, it is judged as no target. Generally, the signal is sent by the signal source, is subject to various interferences during the propagation process, and is processed after reaching the receiver and output to the detector, and then the detector makes a judgment on the input signal according to appropriate criteria.

[0041] In actual radar detection, in order to obtain stable detection performance and a constant false alarm probability, the background clutter power level is often estimated from the echo data measured in real time, so as to adaptively adjust the detection threshold. The key to CFAR processing is to estimate the background clutter power. The main purpose of CFAR technology based on ordered statistics (OS) is to suppress the performance degradation caused by the shielding effect. The reference unit values ​​are sorted, and the kth element of the sorted sequence is called the kth ordered statistic, and the value of this statistic is used as the clutter power estimate.

[0042] 6. Doppler effect

[0043] The Doppler effect refers to the change in the wavelength of radiation emitted by an object due to the relative motion between the wave source and the observer. In front of the moving wave source, the wave is compressed, the wavelength becomes shorter, and the frequency becomes higher, which is called blue shift; behind the moving wave source, the opposite effect occurs. The wavelength becomes longer and the frequency becomes lower, which is called red shift; the higher the speed of the wave source, the greater the effect. Based on the degree of blue shift or red shift of the wave, the speed of the wave source moving in the direction of observation can be calculated.

[0044] The above is an introduction to the technical terms involved in the embodiments of the present application, which will not be repeated below.

[0045] In the field of smart home, it is of great significance to control home appliances by identifying the status of people in bed, which allows home appliances to adjust according to the status of people at any time. For example, taking the home appliance as an air conditioner, when the air conditioner detects that a person is in bed, it means that the person may be in a sleeping state. At this time, the air conditioner can be adjusted to sleep mode. For another example, when the air conditioner detects that a person is not in bed, it means that the person may be in an active state. At this time, the air conditioner can be adjusted to normal mode. In this way, the air conditioner can adjust the working mode according to whether the person is in bed, so as to provide better services to users.

[0046] Among them, the bed status recognition technology is the key link to determine the bed status of a person. Generally, the bed status of a person can be identified by a sensor. For example, the sensor can be a pressure sensor. For example, one or more pressure sensors can be set on the mattress. In this way, the mattress can determine whether there is a person on the mattress based on the pressure values ​​detected by the one or more pressure sensors. The sensor has high recognition accuracy, but the cost is high and it is difficult to integrate other functions.

[0047] In view of this, the present application provides a state recognition method, which includes: obtaining multiple frames of echo signals within a detection period through a radar device; processing the multiple frames of echo signals within the detection period to obtain multiple Doppler data of the detection period. Then, determining whether there is a target human body in the target area based on the multiple Doppler data, and when it is determined that there is a target human body, determining the position information of the target human body based on the multiple frames of echo signals. Finally, the position information of the bed and the position information of the target human body are input into a preset state detection model to obtain a recognition result of whether the target human body is in the bed state.

[0048] The state recognition method provided in the embodiment of the present application can determine whether there is a target human body in the target area according to the multi-frame echo signal within a preset period acquired by the radar device, and when it is determined that there is a target human body, determine whether the target human body is in bed according to the preset state detection model. Without using a shooting device, the bed status of users in the area can be monitored in real time while protecting the user's personal privacy. In addition, the radar device will not be affected by environmental issues such as light, so this method is more practical and universal.

[0049] In addition, the radar equipment mainly uses the reflected signals of electromagnetic waves from people to detect people. While identifying the bed status, it can also integrate other functional modules to serve users.

[0050] The embodiment of the present application also provides a state recognition device, which can be used to execute the above-mentioned state recognition method. Optionally, the state recognition device can be a radar device, or the state recognition device can also be a server, a smart home device or other device with data processing capabilities, which is connected to the radar device so that data can be transmitted between the state recognition device and the radar device.

[0051] In one example, the state recognition device is a radar device, for example, and the radar device can communicate data with the air conditioner. Figure 1As shown, when the air conditioner is in the on state, the air conditioner can send a bed status recognition instruction to the radar device. After receiving the bed status recognition instruction, the radar device can transmit a detection signal to the area where it is located. After receiving the echo signal, the radar device can perform signal processing, and perform state recognition based on the processed signal to obtain a state recognition result. When the state recognition result is that there is no person in the area, a response message that there is no person in the area can be sent to the air conditioning device, so that after receiving the response message, the air conditioning device can perform corresponding actions according to the preset instructions. When the state recognition result is that there is a person in the area, the radar device can further perform bed status recognition and send the recognition result to the air conditioning device, so that the air conditioning device can perform corresponding actions according to the recognition result.

[0052] The radar device is an electronic device that uses electromagnetic waves to detect targets, such as millimeter wave radar, microwave radar, ultra-wideband radar, etc. In some embodiments of the present application, a millimeter wave radar with strong anti-interference ability, strong resolution ability and high measurement accuracy is used.

[0053] In addition, the radar device can be deployed at a reasonable location in the home to facilitate its detection of the target human body. Optionally, the radar device can be deployed on home devices, such as smart air conditioners, smart TVs and other devices.

[0054] The radar device can be in the same plane as the target body, or in a different plane from the target body. For example, when the radar device is installed in front of the bed, if the target body is located on the bed, the radar device and the target body are in the same plane. For another example, when the radar device is installed above the bed (such as on a wall), the radar device and the target body can be in different planes.

[0055] For example, taking the state recognition device as a server, the state recognition device may be a single server, or may be a server cluster composed of multiple servers. In some implementations, the server cluster may also be a distributed cluster.

[0056] In the embodiment of the present application, the server can be connected to the radar device, and the radar device can collect echo data in its detection space in real time and send the collected echo data to the server. Correspondingly, the server can receive the echo data collected by the radar device, and execute the above-mentioned state recognition method according to the received echo data to obtain the state recognition result of the target human body.

[0057] Optionally, the server can also be connected to a smart home such as a smart air conditioner, or the server can be connected to a smart home system. Thus, the server can control smart homes such as a smart air conditioner in real time according to the state recognition result of the target human body, thereby providing a comfortable environment for the target human body.

[0058] like Figure 2 As shown, a state identification method is provided for an embodiment of the present application, and the method comprises the following steps:

[0059] S201, acquiring multiple frames of first echo signals of a target area within a detection period.

[0060] The target area may refer to a closed space area, and objects in the area may reflect the signal emitted by the radar device. For example, the target area may be a room. Thus, after the radar device emits a detection signal, objects in the room (such as a table, a wall, a bed, a person, etc.) may reflect the detection signal back. Thus, the radar device may receive the reflected first echo signals of multiple frames.

[0061] In actual usage scenarios, radar equipment usually periodically emits electromagnetic waves at a preset frequency and obtains echo signals reflected in the target area. Exemplarily, the radar equipment may include multiple signal receiving channels, and a detection cycle may include multiple frames of echo signals. Then each frame of echo signals may include radar signals of multiple cycles and multiple channels received by the radar equipment. For example, the radar equipment may be provided with 12 channels. That is, the radar equipment may be provided with 3 transmitting antennas and 4 receiving antennas. The multiple receiving antennas may be sorted in a certain order. For example, the 4 receiving antennas may be arranged horizontally or in two rows (one row includes 2).

[0062] The duration of the detection cycle can be set as needed, for example, it can be 3 seconds or 2 seconds without limitation.

[0063] S202: Process multiple frames of first echo signals within a detection period to obtain multiple frames of processed first echo signals.

[0064] Processing the first echo signals of multiple frames can be used to eliminate clutter signals in the first echo signals of multiple frames. The processing method may include pulse compression, MTI, incoherent accumulation, etc. The signal-to-noise ratio of the processed first echo signal is greater than the signal-to-noise ratio of the first echo signal before processing.

[0065] In one example, Figure 3As shown, after the state recognition device obtains multiple frames of first echo signals within the detection period through the radar device, it first performs pulse compression on each frame of the first echo signal to obtain a narrow pulse signal, so as to facilitate subsequent processing of the echo signal. Then, the state recognition device can process each frame of narrow pulse signal using the moving target display technology to eliminate the interference and influence of clutter signals. Furthermore, after eliminating clutter signals in the multiple frames of first echo signals, the state recognition device can also perform incoherent accumulation on the multiple frames of echo signals to obtain an echo signal with a higher signal-to-noise ratio.

[0066] When the state recognition device detects that there is a target person in the target area, the state recognition device can identify the target person's bed status and report the target person's bed status; when the state recognition device detects that there is no target person in the target area, it can output the recognition result as "no one". Then, the air conditioning equipment can perform corresponding operations according to the recognition result.

[0067] In another example, Figure 4 As shown, when the radar device is provided with 12 channels, after receiving the echo signals of the 12 channels, the state recognition device can perform pulse compression, clutter suppression and incoherent accumulation on the echo signals to obtain processed echo signals.

[0068] It should be noted that, affected by the objects in the target area, the echo signals obtained by the state recognition device when the radar device detects the target area may include clutter signals generated by a large number of bedroom furnishings such as beds and wardrobes, which will interfere with the subsequent analysis of the echo signals generated by the target human body. Therefore, the state recognition device can first pre-process the obtained echo signals to remove the clutter signals generated by the stationary targets in the obtained background.

[0069] Furthermore, the state recognition device determines whether there is a target human body in the target area according to the processed multi-frame first echo signals.

[0070] In one example, the state recognition device determines a plurality of Doppler data in the target area according to the processed multi-frame first echo signals, and determines whether the target area exists according to the plurality of Doppler data.

[0071] Wherein, one Doppler data corresponds to a frame of the first echo signal. The multiple Doppler data can be used to reflect whether there is a detection target (i.e., the target human body in the embodiment of the present application) within the detection time of the echo signal. For example, when the multiple Doppler data are 0, it means that there is no target human body in the target area. When the multiple Doppler data have data that is not 0, it means that there is a target human body in the target area.

[0072] Specifically, the state recognition device can process multiple frames of echo signals within the detection period. Since each frame of echo signals can include radar signals of multiple periods and multiple channels received by the radar device, the state recognition device can perform FFT in the slow time dimension for each frame of echo signals within the detection period, and can obtain multiple Doppler data within the detection period.

[0073] In another example, the state recognition device may use an OS CFAR-based detection method to detect the processed multi-frame first echo information to determine whether there is a target human body in the target area. For details, reference may be made to the prior art and will not be described in detail.

[0074] S203: When it is determined that a target human body exists in the target area, the position information of the target human body is determined according to the processed multi-frame first echo signals.

[0075] The target human body refers to one or more users located in the target area. The position information of the target human body may include the distance and azimuth angle between the target human body and the radar device. The calculation method of the distance and azimuth angle between the target human body and the radar device can refer to the description of the following embodiment, which is not repeated here.

[0076] In one example, for each frame of the processed multiple frames of the first echo signal, the state recognition device can determine multiple point cloud data corresponding to the first echo signal, and determine the position information of the center point of the first echo signal based on the multiple point cloud data. In this way, after determining the position information of the center point of each frame of the first echo signal, the state recognition device can determine the position information of the target human body based on the position information of the center point of the multiple frames of the first echo signal.

[0077] Among them, a point cloud data includes a distance value and an azimuth angle of a signal reflection point of the target human body. The distance value of the signal reflection point refers to the distance between the signal reflection point and the antenna (i.e., the receiving antenna) that receives the first echo signal. The azimuth angle of the signal reflection point can refer to the angle between the transmission direction of the first echo signal and the receiving antenna. For example, Figure 5 As shown, it is the azimuth angle of the signal reflection point provided in the embodiment of the present application.

[0078] Among them, the center point of the first echo signal can be used to characterize the target human body. The center point can reflect the main position of the target human body. The position information of the center point may include a target distance value and a target azimuth angle. The target distance value can be determined based on the distance values ​​of multiple signal reflection points, for example, it can be an average value or a median value of the distance values ​​of multiple signal reflection points. The target azimuth angle can be determined based on the azimuth angles of multiple signal reflection points, for example, it can be an average value or a median value of the azimuth angles of multiple signal reflection points.

[0079] It should be noted that when there is a target human body in the target area, since the distances between different reflection points of the target human body and the radar device are different, the time for the echo signals emitted from different reflection points to be transmitted to the radar device is also different. At this time, the radar device can regard multiple echo signals received at the same time as echo signals reflected from the same reflection point. That is to say, within one cycle, when the radar device emits a detection signal once, the radar device can receive different echo signals at different times. The different echo signals can be echo signals reflected from different reflection points of the target human body.

[0080] Furthermore, when the radar device has multiple receiving antennas, since the difference in the distance values ​​between the multiple receiving antennas and the same reflection point is small, the time difference between the multiple receiving antennas receiving the echo signals reflected from the same reflection point is also small. Based on this, the state recognition device can regard the multiple echo signals received at the same time as echo signals reflected from the same reflection point.

[0081] S204: Acquire the position information of the bed in the target area.

[0082] The position information of the bed may include the distance and azimuth angle between the bed and the radar device.

[0083] In a possible implementation, the state recognition device may store the position information of the bed in advance, and the state recognition device may also obtain the position information of the bed in the target area from other devices, for example, the position information of the bed in the target area may be obtained from a server. In this way, when the state recognition device recognizes the in-bed state of the target person, it can quickly obtain the position information of the bed, which is simple and convenient.

[0084] In another possible implementation, after a household appliance (such as an air conditioner) equipped with a state recognition device is installed in the target area, in response to an initial power-on command, the state recognition device can emit a detection signal and determine the position information of the bed based on the received second echo signal.

[0085] The first power-on instruction may be a power-on instruction sent by the remote controller to the home appliance in response to the user touching or pressing the power-on button for the first time. During the first operation of the home appliance, the target person sits on the bed.

[0086] In one example, the state recognition device can output a first prompt message in response to an initial power-on command, and the first prompt message is used to prompt the user to sit on the bed (such as sitting in the center of the bed). In response to the instruction to determine that the user is sitting on the bed, the state recognition device transmits multiple frames of detection signals through a radar device. After the state recognition device receives multiple frames of second echo signals, it can determine the position information of the target human body based on the multiple frames of second echo signals, and use the position information of the target human body as the position information of the bed. In this way, the state recognition device can accurately determine the position information of the bed.

[0087] For example, Figure 6 As shown, the state recognition device can determine multiple point cloud data of each frame of the second echo signal in multiple frames of the second echo signal. For each frame of the second echo signal, the state recognition device can determine the position information of the corresponding center point based on the multiple point cloud data of the second echo signal. In this way, the state recognition device can determine the position information of multiple center points. Then, the state recognition device counts the position information of the multiple center points according to the two-dimensional histogram statistics method to determine the position information of the target human body, that is, the position information of the bed. The two-dimensional histogram statistics method can refer to Figure 7 Description.

[0088] It should be noted that in order to avoid the problem that the working state of the home is inconsistent with the user's state in bed after the position information of the bed is changed. The state recognition device can execute the above second possible implementation method according to a preset period. The preset period can be one month, three months, etc. Alternatively, when the state recognition device determines that the target human body exists in the target area and the existence time is in a preset time period (such as 9 pm to 6 am the next day, that is, the sleeping time period), when the state recognition device detects that the target human body is not located in the bed, it means that the position information of the bed may have changed. When it is determined that the position information of the bed has changed, the state recognition device can re-execute the above second possible implementation method. In this way, the problem of errors in the recognition results can be avoided.

[0089] S205: Determine the in-bed state of the target person according to the position information of the bed, the position information of the target person, and a preset state monitoring model.

[0090] Among them, the preset state detection model is used to detect the in-bed state of the target person. The input of the preset state detection model can be the position information of the bed and the position information of the target person, or it can also be information processed according to the position information of the bed and the position information of the target person. The output of the preset state detection model can be the in-bed state of the target person. The preset state detection model can be pre-configured by the state recognition device, or it can be pre-trained, without limitation. The training method of the preset state detection model can refer to the subsequent description and will not be repeated here.

[0091] It should be noted that, in the embodiments of the present application, the target human body may refer to one user or multiple users. When the target human body is one user, the in-bed status of the target human body includes the target human body in bed and the target human body not in bed. The target human body in bed means that the position information of the target human body is consistent with the position information of the bed or the distance between the target human body and the bed does not exceed a preset value. The target human body not in bed means that the distance between the target human body and the bed exceeds a preset value. When the target human body is multiple users, the target human body in bed may mean that the multiple users are all in bed; the target human body not in bed may mean that at least one of the multiple users is not in bed, and the other users may be in bed or not in bed.

[0092] In a possible implementation, the state recognition device may first process the position information of the target person and the position information of the bed, and input the processed information into a preset state detection model to obtain the in-bed state of the target person.

[0093] Specifically, the state recognition device may first determine a plurality of characteristic values ​​according to the position information of the target human body and the position information of the bed, and input the plurality of characteristic values ​​into a preset state detection model to obtain the in-bed state of the target human body.

[0094] The multiple eigenvalues ​​may be used to characterize the deviation between the position information of the target human body and the position information of the bed. For example, the multiple eigenvalues ​​may include a first eigenvalue and a second eigenvalue. The first eigenvalue may be the difference between the position information of the target human body and the position information of the bed (such as the absolute value of the difference between the position information of the target human body and the position information of the bed), and the second eigenvalue may be the offset value between the position information of multiple center points of the target human body.

[0095] In one example, the offset values ​​between the position information of multiple center points may satisfy Formula 1.

[0096]

[0097] Among them, P d Identifies the offset value between the position information of multiple center points of the target human body. N represents the number of multiple center points of the target human body, and N is a positive integer.n ,y n ) represents the location information of the nth center point. n is the distance value of the nth center point, y n is the azimuth angle of the nth center point.

[0098] based on Figure 2 According to the technical solution shown, in the embodiment of the present application, the state recognition device can pre-process the received multi-frame echo information to obtain a multi-frame echo signal with a high signal-to-noise ratio. In this way, the interference of noise on the echo signal can be reduced. Then, the state recognition device can determine whether there is a target human body in the target area based on the processed multi-frame echo signal. When it is determined that there is a target human body in the target area, the position information of the target human body can be determined based on the processed multi-frame echo signal. Since the signal-to-noise ratio of the processed multi-frame echo signal is high, the state recognition device can more accurately determine the position information of the target human body based on the multi-frame echo signal with a high signal-to-noise ratio. Finally, after determining the position information of the bed, the state recognition device can determine the position information of the target human body based on the position information of the target human body, the position information of the bed and the preset state detection model. Since the preset state detection model can be used to determine the in-bed state of the target human body, therefore, in the embodiment of the present application, the in-bed state of the target human body can be accurately determined by the preset state model.

[0099] In a possible embodiment, in the above S203, the method for calculating the distance and azimuth angle between the target human body and the radar device may be as follows:

[0100] 1. Method for calculating the distance between the target human body and the radar equipment.

[0101] In one example, the state recognition device may be provided with a timer. When the state recognition device sends a detection signal, the timer is triggered to start timing; when the state recognition device receives an echo signal, the timer is triggered to stop timing. In this way, the state recognition device can calculate the distance value of the signal reflection point according to the time recorded by the timer and the transmission rate of the signal.

[0102] 2. The calculation method of the azimuth angle between the target human body and the radar equipment.

[0103] In one example, the state recognition device may calculate the azimuth angle of the signal reflection point according to the distance between the multiple receiving antennas and the echo signals received by the multiple receiving antennas.

[0104] It should be noted that, due to the different propagation distances of the echo signal from the reflection point to the receiving antenna, the phases of the echo signals received by different receiving antennas are also different. Figure 4The state recognition device calculates the azimuth angle of the signal reflection point according to the phase difference between the echo signal received by the receiving antenna Rx1 and the echo signal received by the receiving antenna Rx2 and the distance between the two receiving antennas. When the distance between the two receiving antennas is less than the wavelength of the echo signal, the azimuth angle of the signal reflection point can be calculated according to Formula 2.

[0105]

[0106] Here, θ represents the azimuth angle. Represents the phase difference between the echo signals received by the two receiving antennas. d represents the distance between the two receiving antennas. λ represents the wavelength of the echo signal.

[0107] It should be noted that when the target body is in motion, the point cloud data of the target body is distributed in pieces; when the target body is stationary, some parts of the target body may be stationary, and the echo signals reflected by the stationary parts may be identified as other objects, resulting in the point cloud data of the target body being missed. Therefore, the center point of the target body can be determined based on the multiple point cloud data. Then, the position information of the target body is determined based on the center point of the target body corresponding to the multiple frames of echo signals.

[0108] In one possible implementation, Figure 7 As shown, the method for the state recognition device to determine the position information of the target human body according to the position information of the center point of the multi-frame first echo signal may specifically include S701 to S703.

[0109] S701: Divide the target area into multiple sub-areas.

[0110] Among them, a sub-region has corresponding location information.

[0111] In one example, the state recognition device can be divided according to the field of view of the radar device to obtain multiple sub-areas. The field of view of the radar device can be a plane, and each sub-area can be a cell of a preset size. For example, the preset size can be 0.1 meters * 0.1 meters, without limitation.

[0112] S702: Count the number of central points included in each of the multiple sub-regions.

[0113] The state recognition device may compare the position information of the center point with the position information of each sub-region, thereby determining the sub-region corresponding to each center point. When determining the sum of the number of center points included in each sub-region, the state recognition device may count the number of center points included in each sub-region.

[0114] S703: Use the position information of the sub-region with the largest number of center points among the multiple sub-regions as the position information of the target human body.

[0115] Based on the above S701 to S703, when the state recognition device determines the position information of the target body according to the position information of multiple sub-areas, the more center points a sub-area includes, the greater the probability that the target body appears in the sub-area. Based on this principle, the position information of the sub-area with the largest number of center points is used as the position information of the target body, which is accurate and simple.

[0116] The above describes the process of determining the target person's bed state according to the preset state detection model. The following takes the execution subject as the state detection model as an example to illustrate the training process of the preset state detection model.

[0117] In a possible embodiment, Figure 8 As shown, the training process of the preset state detection model may include S801 to S805.

[0118] S801. Acquire multiple frames of echo signals in a preset time period, and process the multiple frames of echo signals to obtain processed echo signals.

[0119] The preset time period may include one or more first sub-time periods and one or more second sub-time periods. In the first sub-time period, the target human body is on the bed. In the second sub-time period, the target human body is not on the bed. One frame of echo signal corresponds to one sub-time period. Thus, the multi-frame echo signal may include one or more frames of echo signal when the target human body is on the bed, and one or more frames of echo signal when the target human body is not on the bed.

[0120] The method for processing the multi-frame echo signals may refer to the above-mentioned S202 and will not be described in detail.

[0121] S802: Determine the position information of the target human body corresponding to each frame of the echo signal according to the processed multiple frames of echo signals.

[0122] The implementation method of S802 may refer to the above S203 and will not be described in detail.

[0123] S803: Obtain the position information of the bed.

[0124] The implementation method of S803 may refer to the above S204 and will not be described in detail.

[0125] S804: Process the position information of the bed and the position information of multiple target human bodies to obtain a first eigenvalue set and a second eigenvalue set.

[0126] The first eigenvalue set includes a plurality of first eigenvalues, and the second eigenvalue set includes a plurality of second eigenvalues.

[0127] In one example, the state recognition device may substitute the position information of the target human body and the position information of the bed corresponding to each frame of the echo signal into the following formula 3 and formula 4 to obtain the first eigenvalue and the second eigenvalue of the frame of the echo signal.

[0128] P err (n)=|P(n)-P bed | Formula 3

[0129]

[0130] Among them, P err P(n) represents the difference between the position information of the target human body corresponding to the n-th frame echo signal and the position information of the bed, that is, the first eigenvalue. P(n) represents the position information of the target human body corresponding to the n-th frame echo signal. bed Indicates the position information of the bed. d (n) represents the offset value corresponding to the n-th frame echo signal, that is, the second eigenvalue.

[0131] Exemplarily, the first eigenvalue set may include: P err (1) P err (2), ..., P err (n). The first feature value set may include: P d (1) P d (2), ..., P d (n).

[0132] S805: Train the first eigenvalue set, the second eigenvalue set, and the corresponding target human body's bed status to obtain a preset status detection model.

[0133] In one example, the state detection model can be trained on the first eigenvalue set, the second eigenvalue set, and the corresponding bed state of the target human body according to a preset algorithm to obtain a preset state detection model.

[0134] Among them, the preset algorithm can be a support vector machine (SVM) classification algorithm, a neural network algorithm, etc.

[0135] In another example, the state detection model may input the first eigenvalue set, the second eigenvalue set, and the corresponding target human body's in-bed state into a machine learning model for training to obtain a preset state detection model.

[0136] Among them, the machine learning model can be an SVM classifier, etc.

[0137] Based on the above S801 to S805, it can be known that the state recognition device can be trained according to a plurality of sample echo signals and the corresponding target human body's in-bed state to accurately obtain a pre-state detection model.

[0138] The above mainly introduces the solution provided by the present application from the perspective of the method. It is understandable that in order to realize the above functions, the radar device or server includes a hardware structure and / or software module corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the algorithm steps of each example described in the embodiments disclosed herein, the present invention can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0139] The present application can divide the radar device or server into functional modules according to the above method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the present application is schematic and is only a logical functional division. There may be other division methods in actual implementation.

[0140] Fig. 9 FIG. 2 is a schematic diagram showing the composition of a state recognition device provided in an embodiment of the present application. Fig. 9 As shown, the state recognition device 900 includes an acquisition unit 901 and a processing unit 902 .

[0141] The acquisition unit 901 is used to acquire multiple frames of first echo signals within a detection period through a radar device.

[0142] The processing unit 902 is used to process multiple frames of first echo signals within the detection period to obtain multiple frames of processed first echo signals, wherein the signal-to-noise ratio of the processed first echo signals is greater than the signal-to-noise ratio of the first echo signals before processing.

[0143] The processing unit 902 is further configured to determine the position information of the target human body according to the processed multiple frames of first echo signals when it is determined that there is a target human body in the target area.

[0144] The acquisition unit 901 is also used to acquire the position information of the bed in the target area.

[0145] The processing unit 902 is further configured to determine the target person's bed state according to the bed position information, the target person's position information, and a pre-state detection model for detecting the target person's bed state.

[0146] In some embodiments, for each frame of the processed multiple frames of first echo signals, the processing unit 902 is specifically used to: determine multiple point cloud data corresponding to the first echo signal, one point cloud data includes a distance value and an azimuth angle of a signal reflection point of the target human body, the distance value of the signal reflection point is the distance between the signal reflection point and the antenna receiving the first echo signal, and the azimuth angle of the signal reflection point is the angle between the transmission direction of the first echo signal and the normal of the antenna receiving the first echo signal; determine the position information of the center point corresponding to the first echo signal based on the multiple point cloud data, the position information of the center point includes a target distance value and a target azimuth angle, the target distance value is determined based on the distance values ​​of the multiple signal reflection points, and the target azimuth angle is determined based on the azimuth angles of the multiple signal reflection points; determine the position information of the target human body based on the position information of the multiple center points.

[0147] In some embodiments, the processing unit 902 is specifically used to: divide the target area into multiple sub-areas, each sub-area has corresponding position information; count the number of center points included in each of the multiple sub-areas; and use the position information of the sub-area with the largest number of center points among the multiple sub-areas as the position information of the target human body.

[0148] In some embodiments, the processing unit 902 is specifically used to: determine a first eigenvalue and a second eigenvalue according to the position information of the bed and the position information of the target person, the first eigenvalue being the difference between the position information of the bed and the position information of the target person, and the second eigenvalue being the offset value between the position information of multiple center points of the target person; input the first eigenvalue and the second eigenvalue into a preset state detection model to obtain the in-bed state of the target person.

[0149] In some embodiments, the processing unit 902 is specifically used to: obtain multiple frames of second echo signals in the target area within a first preset period, the target human body is located on the bed within the first preset period, and the first preset period is before the detection period; determine the position information of the target human body according to the multiple frames of second echo signals, and use the position information of the target human body as the position information of the bed.

[0150] Fig. 9 The units in a module may also be referred to as modules, for example, a processing unit may be referred to as a processing unit. Fig. 9 In the illustrated embodiment, the names of the various units may not be the names shown in the figure. For example, the acquisition unit may also be called a transceiver unit.

[0151] Fig. 9 If each unit in the embodiment is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform all or part of the steps of the method described in each embodiment of the present application. The storage medium for storing computer software products includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0152] The embodiment of the present application also provides a schematic diagram of the hardware structure of a radar device or server, such as Fig.10 As shown, the server 1000 includes a processor 1001, and optionally, further includes a memory 1002 and a transceiver 1003 connected to the processor 1001. The processor 1001, the memory 1002 and the transceiver 1003 are connected via a bus 1004.

[0153] Processor 1001 may be a central processing unit (CPU), a general-purpose processor network processor (NP), a digital signal processor (DSP), a microprocessor, a microcontroller, a programmable logic device (PLD), or any combination thereof. Processor 1001 may also be any other device having a processing function, such as a circuit, a device, or a software module. Processor 1001 may also include multiple CPUs, and processor 1001 may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. The processor here may refer to one or more devices, circuits, or processing cores for processing data (such as computer program instructions).

[0154] The memory 1002 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, and the embodiments of the present application do not impose any restrictions on this. The memory 1002 may exist independently or be integrated with the processor 1001. Among them, the memory 1002 may contain a computer program code. The processor 1001 is used to execute the computer program code stored in the memory 1002, thereby realizing the method provided in the embodiments of the present application.

[0155] The transceiver 1003 may be used to communicate with other devices or communication networks (such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.) The transceiver 1003 may be a module, a circuit, a transceiver or any device capable of achieving communication.

[0156] The bus 1004 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus 1004 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.10 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0157] The embodiment of the present application also provides a computer-readable storage medium, including computer-executable instructions, which, when executed on a computer, enable the computer to execute any one of the methods provided in the above embodiments.

[0158] The embodiment of the present application also provides a computer program product including computer-executable instructions, which, when executed on a computer, enables the computer to execute any one of the methods provided in the above embodiments.

[0159] An embodiment of the present application also provides a chip, including: a processor and an interface, wherein the processor is coupled to a memory via the interface, and when the processor executes a computer program or a computer execution instruction in the memory, any one of the methods provided in the above embodiments is executed.

[0160] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer-executable instructions. When the computer-executable instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer-executable instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer-executable instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server, data center, etc. that contains one or more servers that can be integrated with the medium. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).

[0161] Although the present application is described herein in conjunction with various embodiments, in the process of implementing the claimed application, those skilled in the art may understand and implement other changes to the disclosed embodiments by viewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "one" or "an" does not exclude multiple situations. A single processor or other unit may implement several functions listed in the claims. Certain measures are recorded in different dependent claims, but this does not mean that these measures cannot be combined to produce good results.

[0162] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely exemplary illustrations of the present application as defined by the appended claims, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

[0163] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A state recognition method, characterized in that: The method comprises: Acquire multiple frames of first echo signals of the target area within a detection period; Processing the multiple frames of first echo signals within the detection period to obtain the multiple frames of processed first echo signals, wherein the signal-to-noise ratio of the processed first echo signals is greater than the signal-to-noise ratio of the first echo signals before the processing; When it is determined that a target human body exists in the target area, determining the position information of the target human body according to the processed multi-frame first echo signals; Acquire the position information of the bed in the target area, and determine a first characteristic value and a second characteristic value according to the position information of the bed and the position information of the target person, wherein the first characteristic value is the difference between the position information of the bed and the position information of the target person, and the second characteristic value is the offset value between the position information of multiple center points of the target person; The first characteristic value and the second characteristic value are input into a preset state detection model to obtain the bed state of the target human body, and the preset state detection model is used to detect the bed state of the target human body.

2. The method according to claim 1, characterized in that Determining the position information of the target human body according to the processed multiple frames of first echo signals includes: For each frame of the processed multiple frames of first echo signals, determine a plurality of point cloud data corresponding to the first echo signal, wherein one point cloud data comprises a distance value and an azimuth angle of a signal reflection point of the target human body, the distance value of the signal reflection point being the distance between the signal reflection point and an antenna receiving the first echo signal, and the azimuth angle of the signal reflection point being the angle between a transmission direction of the first echo signal and a normal of the antenna receiving the first echo signal; Determine, according to the plurality of point cloud data, position information of a center point corresponding to the first echo signal, wherein the position information of the center point includes a target distance value and a target azimuth angle, wherein the target distance value is determined according to the distance values ​​of the plurality of signal reflection points, and the target azimuth angle is determined according to the azimuth angles of the plurality of signal reflection points; The position information of the target human body is determined according to the position information of the multiple center points.

3. The method according to claim 2, characterized in that Determining the position information of the target human body according to the position information of the plurality of center points includes: Dividing the target area into a plurality of sub-areas, each sub-area having corresponding location information; Counting the number of the central points included in each of the multiple sub-areas; The position information of the sub-region including the largest number of the central points among the multiple sub-regions is used as the position information of the target human body.

4. The method according to any one of claims 1 to 2, characterized in that: The determining the position information of the bed in the target area includes: Acquire multiple frames of second echo signals in the target area within a first preset period, wherein the target human body is located on the bed within the first preset period, and the first preset period is before the detection period; The position information of the target human body is determined according to the multiple frames of second echo signals, and the position information of the target human body is used as the position information of the bed.

5. A state recognition device, characterized in that: include: An acquisition unit, used for acquiring a plurality of frames of first echo signals of a target area within a detection period; A processing unit, used for processing the multiple frames of first echo signals within the detection period to obtain the multiple frames of first echo signals after processing, wherein the signal-to-noise ratio of the processed first echo signals is greater than the signal-to-noise ratio of the first echo signals before processing; The processing unit is further configured to determine the position information of the target human body according to the processed multiple frames of first echo signals when it is determined that the target human body exists in the target area; The acquisition unit is further used to acquire the position information of the bed in the target area; The processing unit is further used to determine a first characteristic value and a second characteristic value according to the position information of the bed and the position information of the target person, wherein the first characteristic value is a difference between the position information of the bed and the position information of the target person, and the second characteristic value is an offset value between the position information of multiple center points of the target person; The first characteristic value and the second characteristic value are input into a preset state detection model to obtain the bed state of the target human body, and the preset state detection model is used to detect the bed state of the target human body.

6. The device according to claim 5, characterized in that For each frame of the processed first echo signal in the plurality of frames of the first echo signal, the processing unit is specifically configured to: Determine a plurality of point cloud data corresponding to the first echo signal, wherein one point cloud data includes a distance value and an azimuth angle of a signal reflection point of the target human body, the distance value of the signal reflection point is the distance between the signal reflection point and an antenna receiving the first echo signal, and the azimuth angle of the signal reflection point is the angle between a transmission direction of the first echo signal and a normal of the antenna receiving the first echo signal; Determine, according to the plurality of point cloud data, position information of a center point corresponding to the first echo signal, wherein the position information of the center point includes a target distance value and a target azimuth angle, wherein the target distance value is determined according to the distance values ​​of the plurality of signal reflection points, and the target azimuth angle is determined according to the azimuth angles of the plurality of signal reflection points; The position information of the target human body is determined according to the position information of the multiple center points.

7. The device according to claim 6, characterized in that The processing unit is specifically used for: Dividing the target area into a plurality of sub-areas, each sub-area having corresponding location information; Counting the number of the central points included in each of the multiple sub-areas; The position information of the sub-region including the largest number of the central points among the multiple sub-regions is used as the position information of the target human body.

8. A state recognition device, characterized in that: The state recognition device comprises: at least one processor; at least one storage device; The at least one memory stores computer instructions, and when the computer instructions are executed by the processor, the state recognition device executes the state recognition method according to any one of claims 1 to 4.

Images