Sitting and standing conversion information determination method and device and electronic equipment

Abstract

The embodiment of the invention discloses a sitting-standing conversion information determination method and device and electronic equipment. The method comprises the following steps: based on a received radar echo signal, determining a target motion direction when a target object executes sitting-standing conversion motion; wherein the radar echo signal is measurement data obtained after radar measurement is carried out on the environment where the target object is located; in a plurality of preset information extraction modes, determining a target information extraction mode corresponding to the target motion direction; and determining sitting and standing conversion information of the target object based on the target information extraction mode and the radar echo signal. According to the technical scheme, the movement direction of the target object does not need to be restrained, diversified movement modes of the target object in actual life are met, and the sitting and standing conversion information of the target object can be effectively and accurately determined in home life.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of information detection technology, and more particularly to a method, device, and electronic device for determining sit-stand transition information. Background Art

[0002] Sit-to-stand transition estimation is an important human behavior estimation task with broad applications in medical rehabilitation, elderly care, identity recognition, and home monitoring. Accurately capturing stand-up motion data is crucial for assessing lower limb strength, monitoring rehabilitation training outcomes, and preventing falls.

[0003] Existing technologies use radar measurement to determine the sit-to-stand transition information of the measured subject. However, effective determination of this information requires the measured subject to be in a preset environment and to move according to pre-defined motion requirements. This motion requirement restricts the measured subject's freedom of movement, making it difficult to adapt to the diverse motion patterns of real-life users and preventing effective and accurate determination of the measured subject's sit-to-stand transition information at home. Summary of the Invention

[0004] The embodiments of the present invention provide a method, device, and electronic device for determining sit-to-stand transition information to meet the diverse motion patterns of a target object in real life and achieve the purpose of effectively and accurately determining the sit-to-stand transition information of a target object in home life.

[0005] According to one aspect of the present invention, a method for determining sit-to-stand transition information is provided, comprising:

[0006] Determining a target movement direction of a target object when performing a sit-to-stand transition movement based on a received radar echo signal, wherein the radar echo signal is measurement data obtained by performing radar measurement of an environment in which the target object is located;

[0007] Determining a target information extraction method corresponding to the target motion direction from among a plurality of pre-set information extraction methods;

[0008] Based on the target information extraction method and the radar echo signal, sit-stand conversion information of the target object is determined.

[0009] According to another aspect of the present invention, there is provided a device for determining sit-to-stand transition information, comprising:

[0010] a motion direction determination module, configured to determine the target motion direction of the target object when performing a sit-to-stand transition motion based on a received radar echo signal; wherein the radar echo signal is measurement data obtained by performing radar measurement of the target object's environment;

[0011] An extraction mode determination module, configured to determine a target information extraction mode corresponding to the target motion direction from among a plurality of pre-set information extraction modes;

[0012] An information determination module is used to determine the sitting-standing conversion information of the target object based on the target information extraction method and the radar echo signal.

[0013] According to another aspect of the present invention, an electronic device is provided, comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to perform the sit-to-stand transition information determination method according to any embodiment of the present invention.

[0017] The technical solution of an embodiment of the present invention determines the target motion direction of a target object when performing a sit-to-stand transition based on a received radar echo signal. The radar echo signal is measurement data obtained by performing radar measurement of the target object's environment. Furthermore, a target information extraction method corresponding to the target motion direction is determined from a plurality of pre-set information extraction methods. This allows for flexible determination of the corresponding information extraction method based on the target object's motion direction, without requiring the target object to move in a pre-set motion direction. Furthermore, the target object's sit-to-stand transition information is determined based on the target information extraction method and the radar echo signal. The technical solution of this embodiment eliminates the need to constrain the target object's motion direction, accommodates the diverse motion patterns of the target object in real life, and can effectively and accurately determine the target object's sit-to-stand transition information in daily life.

[0018] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1is a flow chart of a method for determining sit-to-stand transition information according to an embodiment of the present invention;

[0021] Figure 2 is a schematic diagram of the positional relationship between a radar and a target object provided by an embodiment of the present invention;

[0022] Figure 3 is a flow chart of another method for determining sit-to-stand transition information according to an embodiment of the present invention;

[0023] Figure 4 is a flow chart of sit-to-stand transition estimation provided according to an embodiment of the present invention;

[0024] Figure 5 A micro-Doppler time map provided by an embodiment of the present invention;

[0025] Figure 6 2 is a schematic structural diagram of a device for determining sit-stand transition information according to an embodiment of the present invention;

[0026] Figure 7 2 is a schematic structural diagram of an electronic device for implementing the method for determining sit-to-stand conversion information according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "etc." and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] It should be noted that the collection, collection, updating, analysis, processing, use, transmission, and storage of user personal information involved in the technical solutions disclosed herein comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken with respect to user personal information to prevent unauthorized access to user personal information data and maintain the security of user personal information and network security.

[0030] Before introducing the technical solution, we will first provide an example application scenario. This technical solution can be applied to scenarios where sit-to-stand transition information is generated during the sit-to-stand transition process of a target subject moving from a sitting position to a standing position. This sit-to-stand transition information includes speed information and the target subject's posture information. By determining this sit-to-stand transition information, the target subject's mobility can be assessed. For example, the target subject can be an elderly person or a child. By using this sit-to-stand transition information, the mobility of the elderly and children can be assessed.

[0031] In the prior art, the activity information of the target object is usually obtained by adding sensors or radars. When using sensors for measurement, the target object needs to carry it with him and needs to be charged, which makes it inconvenient for the target object to move. Exemplarily, the sensor may include at least one of an inertial sensor, a visual sensor, a pressure sensor, and a speed sensor. The use of radar to obtain the activity information of the target object in the prior art is a non-contact method. Compared with the measurement method of an acceleration sensor, it does not need to be worn by the target object and does not need to be charged. However, the target object is required to move according to the set movement requirements, such as according to the preset movement direction. If it does not move according to the preset movement direction, it will cause errors and affect the accuracy of the information determination. Moving according to the movement requirements will affect the freedom of movement of the target object, resulting in a poor user experience and is not suitable for actual home life.

[0032] Through this technical solution, the corresponding information extraction method can be determined according to the target object's actual movement direction in home life, so that the sit-stand conversion information can be determined from the radar echo signal according to the determined information extraction method, and there is no need to restrict the target object's movement direction. It can be applicable to the target object's home life scene and can still accurately and effectively determine the sit-stand conversion information in actual home life.

[0033] Figure 1 This is a flowchart of a method for determining sit-to-stand transition information according to an embodiment of the present invention. This embodiment is applicable to determining sit-to-stand transition information generated when a target subject transitions from a sitting position to a standing position. This method can be performed by a sit-to-stand transition information determination device, which can be implemented in hardware and / or software.

[0034] like Figure 1As shown, the method of this embodiment may specifically include:

[0035] S110 : Determine a target movement direction of the target object when performing a sit-to-stand transition movement based on the received radar echo signal.

[0036] The radar echo signal is the measurement data obtained by performing radar measurements of the target object's environment. The radar can be a millimeter-wave radar. For example, a 3-transmitter, 4-receiver radar can be selected and installed on one side of the indoor space. This radar can be used to process and transmit radio frequency signals. After the radar is powered on, it collects data indoors in frames until it is powered off. The target object is the object for which sit-to-stand transition information is to be detected. Sit-to-stand transition refers to the movement from a sitting position to a standing position, and the target motion direction is the direction of the target object during the sit-to-stand transition. For example, the target motion direction is used to reflect the orientation of the target object during the sit-to-stand transition, and is the angle between the line connecting the radar and the target object.

[0037] In a specific implementation, the radar can transmit a signal and receive a radar echo signal corresponding to the transmitted signal. The radar echo signal can reflect the motion information of the target object. In indoor daily life scenes, the activity states of human targets are diverse, mainly divided into two states: large movement and micro-movement. Among them, large movement includes actions such as walking, running and jumping, and micro-movement includes actions such as standing, sitting and lying. When the target object is in a large movement state, the energy contained in the radar echo signal is high; when the target object is in a micro-movement state, the energy contained in the radar echo signal is low. The first information of the target object in the large movement state and the second information in the micro-movement state fed back in the radar echo signal can be combined to determine the target movement direction of the target object when performing the sit-to-stand transition movement.

[0038] Specifically, a pre-trained deep learning model can be used to determine the target motion direction of the target object during the sit-to-stand transition from the radar echo signal. For example, the deep learning model can be a point cloud classification model. Alternatively, radar point cloud data corresponding to each frame can be determined based on the radar echo signal, and the target motion direction of the target object can be determined by analyzing the radar point cloud data.

[0039] S120 . Determine a target information extraction method corresponding to the target motion direction from among a plurality of pre-set information extraction methods.

[0040] The information extraction method is a method of extracting sitting-standing conversion information from the radar echo signal.

[0041] In this embodiment, different information extraction methods are pre-set, and a reinforcement learning method can be used to perform adaptive selection based on the target motion direction, and determine the target information extraction method corresponding to the target motion direction from multiple information extraction methods.

[0042] Alternatively, in order to improve the efficiency of determining the target information extraction method and simplify the process, the implementation method of determining the target information extraction method corresponding to the target motion direction among the pre-set multiple information extraction methods is as follows: based on the radar echo signal, determine the connection direction corresponding to the connection between the target object and the radar as the measurement direction; determine the angle between the target motion direction and the measurement direction, and determine the target direction type of the target motion direction based on the angle; based on the correspondence between the pre-set direction type and the information extraction method, use the information extraction method corresponding to the target direction type as the target information extraction method.

[0043] In this embodiment, the target object's position can be determined using the radar echo signal, and the direction of the line connecting the target object's position and the radar's position is used as the measurement direction. For example, the target object's position can be used as the starting point, and the radar's position as the end point to determine the direction of the line. The angle between the direction and the target's motion direction is then measured, with different angles corresponding to different direction types. Optionally, the target direction type includes at least one of a facing-towards-radar direction type, a facing-away-from-radar direction type, and a side-on-radar direction type.

[0044] To explain the process of determining the target direction type more clearly, see Figure 2 . The measurement direction is the direction of the line starting from the position P of the target object and ending at the position L of the radar. Turning clockwise from the target movement direction to the measurement direction, an angle is formed. When the angle is less than 60 degrees or greater than 300 degrees, it is a facing radar direction type; the angle is between 60 degrees and 90 degrees, which is a lower right side facing radar direction type; the angle is between 90 degrees and 120 degrees, which is an upper right side facing radar direction type; the angle is between 240 degrees and 270 degrees, which is an upper left side facing radar direction type, and 270 degrees to 300 degrees, which is a lower left side facing radar direction type; the angle is greater than 120 degrees and less than 240 degrees, which is a back-to-back radar direction type, which is divided into six areas. The lower side facing radar direction type, the upper right side facing radar direction type, the upper left side facing radar direction type, and the lower left side facing radar direction type can be collectively referred to as the side facing radar direction type.

[0045] In a specific implementation, different information extraction methods may be preset for different direction types, and the information extraction method corresponding to the target direction type is used as the target information extraction method.

[0046] This embodiment determines the positional relationship between the target's motion direction and the radar's position according to the size of the included angle, and then determines the corresponding information extraction method for different positional relationships, thereby taking into account the influence of the radar position on the sit-stand conversion information extraction process. The information extraction method is determined in combination with the radar position and the target's motion direction, which is conducive to accurately determining the sit-stand conversion information and reducing errors.

[0047] S130: Determine the sit-stand conversion information of the target object based on the target information extraction method and the radar echo signal.

[0048] In this embodiment, the radar echo signal can be processed based on a target information extraction method, and the target object's sitting-standing transition information can be determined based on the processing results. Optionally, a micro-Doppler time map containing the target object's motion information can be generated based on the radar echo signal. The information extraction method is a method for extracting features from the micro-Doppler time map. Based on the target information extraction method, feature extraction can be performed on the micro-Doppler time map corresponding to the radar echo signal, and the target object's sitting-standing transition information can be determined based on the extraction results.

[0049] The technical solution of an embodiment of the present invention determines the target motion direction of a target object when performing a sit-to-stand transition based on a received radar echo signal. The radar echo signal is measurement data obtained by performing radar measurement of the target object's environment. Furthermore, a target information extraction method corresponding to the target motion direction is determined from a plurality of pre-set information extraction methods. This allows for flexible determination of the corresponding information extraction method based on the target object's motion direction, without requiring the target object to move in a pre-set motion direction. Furthermore, the target object's sit-to-stand transition information is determined based on the target information extraction method and the radar echo signal. The technical solution of this embodiment eliminates the need to constrain the target object's motion direction, accommodates the diverse motion patterns of the target object in real life, and can effectively and accurately determine the target object's sit-to-stand transition information in daily life.

[0050] Figure 3 It is a flowchart of another method for determining sit-to-stand conversion information provided by an embodiment of the present invention. Based on the above embodiment, this embodiment optionally determines the target movement direction of the target object when performing a sit-to-stand conversion movement based on the received radar echo signal, including: generating multi-frame point cloud data corresponding to the target object based on the radar echo signal; generating a micro-Doppler array corresponding to each frame of point cloud data based on each frame of point cloud data; determining the micro-cluster point cloud data set corresponding to the balance stage of the target object in the process of performing the sit-to-stand conversion movement based on the micro-Doppler array corresponding to each frame of point cloud data; wherein the balance stage reflects the body leaning forward stage of the target object before converting from sitting to standing; based on the principal component analysis method, the micro-cluster point cloud data set is analyzed, and the target movement direction of the target object when performing the sit-to-stand conversion movement is determined based on the analysis results. Among them, the explanations of the terms that are the same or corresponding to the above embodiments are not repeated here. Such as Figure 3 As shown, the method includes:

[0051] S210 : Generate multiple frames of point cloud data corresponding to the target object based on the radar echo signal, and generate a target micro-Doppler time map corresponding to the target object based on each frame of the point cloud data.

[0052] The horizontal axis of the target micro-Doppler time graph represents time, and the vertical axis represents the Doppler frequency of the target object.

[0053] In this embodiment, point cloud data is used to reflect the spatial position and other attributes of the target object during its motion, such as signal energy, the distance between the target object and the radar, etc. Point cloud data can be obtained by directly processing the radar echo signal.

[0054] In practical applications, since radar echo signals contain clutter, clutter suppression is required to prevent it from affecting the target's motion information. Point cloud data is then determined based on the radar echo signal after clutter suppression. However, due to the relatively low signal energy in micro-motion conditions, traditional clutter suppression methods can easily drown out the clutter signals in these conditions, rendering them undetectable.

[0055] This embodiment proposes clutter suppression by accumulating multiple frames of radar echo signals, thereby highlighting information in micro-motion states with low signal-to-noise ratios, thereby avoiding omission of information in micro-motion states. Using the signals after clutter suppression, multi-frame point cloud data corresponding to the target object is generated.

[0056] Optionally, based on the radar echo signal, multi-frame point cloud data corresponding to the target object is generated, including: performing fast Fourier transform in the fast time dimension on each frame echo signal in the radar echo signal to obtain a first processed signal; for each frame of the first processed signal, using the first processed signal as a starting signal, accumulating the starting signal and the first processed signal of a preset number of frames after the starting signal to obtain a first signal group; performing clutter suppression processing on each frame of the first processed signal, updating the first processed signal based on the processing result corresponding to the first processed signal, performing clutter suppression processing on each first signal group, and updating the first signal group based on the processing result corresponding to the first signal group; obtaining a first signal corresponding to the first processed signal based on the updated first processed signal, the updated first signal group and the slow time dimension fast Fourier transform. Two processed signals, and a second signal group corresponding to the first signal group; for each second processed signal and the second signal group corresponding to the second processed signal, target detection is performed on the second processed signal and the second signal group based on the constant false alarm algorithm to obtain first detection data corresponding to the second processed signal and second detection data corresponding to the second signal group; two-dimensional angle estimation processing is performed on the first detection data and the second detection data to obtain angle data corresponding to the first detection data and angle data corresponding to the second detection data; for each first detection data, the first detection data and the angle data corresponding to the first detection data are used to form first point cloud data; for each second detection data, the second detection data and the angle data corresponding to the second detection data are used to form second point cloud data.

[0057] In order to clearly explain the sit-to-stand conversion estimation process, Figure 4 is a flow chart of sit-to-stand conversion estimation provided according to an embodiment of the present invention. Figure 4 As shown in Figure 2, the sitting-standing transformation estimation process includes several steps: indoor 5D point cloud imaging, point cloud extraction based on target tracking, and unconstrained sitting-standing transformation estimation.

[0058] In a specific implementation, indoor 5D point cloud imaging includes preprocessing the radar echo signal. Each frame of the radar echo signal can first be subjected to a fast time-dimensional fast Fourier transform to obtain a first processed signal. Through the first processed signal, the information of the target object in a large-motion state with a high signal-to-noise ratio can be effectively reflected. In addition, for each frame of the first processed signal, the first processed signal is used as a starting signal, and the starting signal and the first processed signal of a preset number of frames after the starting signal are accumulated to obtain a first signal group. The preset number of frames is an integer greater than 1, and those skilled in the art can determine the specific value of the preset number of frames according to the actual application. Through the first signal group, information in a micro-motion state with a low signal-to-noise ratio can be highlighted.

[0059] To reduce the impact of clutter on information, clutter suppression can be performed on each frame of the first processed signal, and the first processed signal can be updated based on the processing results corresponding to the first processed signal. Clutter suppression can also be performed on each first signal group, and the first signal group can be updated based on the processing results corresponding to the first signal group. After performing a fast time-dimensional fast Fourier transform, clutter suppression is performed on the first processed signal and the first signal group, respectively. This not only preserves information in large motion states, but also avoids missing information in micro-motion states.

[0060] Furthermore, based on the updated first processed signal, the updated first signal group, and the slow-time fast Fourier transform, a second processed signal corresponding to the first processed signal and a second signal group corresponding to the first signal group are obtained. The second processed signal and the second signal group each include at least one of the following: distance, Doppler frequency, and antenna array index.

[0061] To effectively identify the target object and suppress false alarms, a constant false alarm rate (CFAR) algorithm can be used to perform target detection on the second processed signal and the second signal group, respectively, to obtain first detection data corresponding to the second processed signal and second detection data corresponding to the second signal group. The first detection data and the second detection data each include at least one of the following: distance to the target object, Doppler frequency, and energy information. Spatial beamforming technology is used to perform two-dimensional angle estimation on the first detection data and the second detection data, respectively, to obtain angle data corresponding to the first detection data and angle data corresponding to the second detection data. The first detection data and the second detection data each include the distance between the target object and the radar, the Doppler frequency data of the target object relative to the radar, and signal energy data. The angle data includes azimuth and elevation angles.

[0062] For each first detection data point, the first detection data point and the corresponding angle data form the first point cloud data. For each second detection data point, the second detection data point and the corresponding angle data form the second point cloud data. Therefore, both the first and second point cloud data points are 5D point clouds containing information in five dimensions: range, Doppler, azimuth, pitch, and energy. This generates a 5D point cloud for the interior.

[0063] Furthermore, to obtain more intuitive spatial information, the first and second point cloud data are converted to a rectangular coordinate system through coordinate transformation, resulting in an indoor space point cloud P(x, y, z, v, e). Here, x and y represent the two-dimensional position information of the point cloud, z represents the height information of the point cloud, v represents the velocity information of the point cloud, and e represents the energy information of the point cloud.

[0064] This embodiment can suppress clutter while avoiding the submergence of information in the micro-motion state; moreover, it can not only obtain the azimuth and pitch angles, but also retain the Doppler information and energy information of each point cloud, thereby comprehensively acquiring the motion information of the target object.

[0065] In this embodiment, after obtaining the first point cloud data and the second point cloud data, the following operations may be performed: Figure 4 The second step in the process is point cloud extraction based on target tracking. This step includes tracking prediction, data association, point cloud clustering, and Kalman filtering. Specifically, the target object can be tracked based on the first and second point cloud data, and its motion can be tracked and predicted. This can reduce the effects of false alarms and missed alarms caused by the limitations of the detection algorithm during target detection. It can also correct relevant parameters of the point cloud, such as speed or distance, to establish a stable target tracking trajectory. After target tracking, data association is performed, and finally, Kalman filtering can be performed.

[0066] In a specific implementation, the target tracking process is performed in a single frame. Specifically, for the first point cloud data corresponding to each frame, it can be determined whether a tracking body exists in the first point cloud data and the corresponding second point cloud data. If so, tracking prediction is performed. The prediction stage mainly predicts the position and speed of the target object at the current moment based on the state estimation value and state equation of the target object at the previous moment, and outputs the predicted trajectory point. If there is no tracking body in the previous frame, the prediction stage and data association are skipped and the next frame is directly entered. If there is a tracking body in the previous frame, the tracking body of the current frame is predicted, and the tracking body prediction is performed to obtain the tracking trajectory of the target object. Then, the data association step is executed.

[0067] The data association step consists of two stages. The first stage is to set a tracking gate centered on the predicted trajectory point. The tracking gate can be rectangular, elliptical, or fan-shaped. The point cloud data falling within the tracking gate can be used in the subsequent data association process, thereby screening out the point cloud data corresponding to the target object for association. The second stage is the data association stage. Specifically, multiple first point cloud data and second point cloud data are associated and matched with the predicted trajectory point. Commonly used data association algorithms include at least one of the nearest neighbor data association algorithm, probabilistic data association algorithm, global nearest neighbor data association, and group association.

[0068] Furthermore, for point cloud data that has not been associated, a clustering algorithm can be used to cluster the point clouds. Based on the clustering results, the existence of a tracking volume can be re-determined. If the point cloud data corresponding to several consecutive frames are successfully clustered, a tracking volume is established, with the average point cloud coordinates used as the starting position of the track and the average point cloud Doppler as the starting velocity. If no point cloud data for a preset number of frames is successfully clustered, the point cloud data can be deleted. Clustering algorithms include DBSCAN (Density-Based Spatial Clustering of Applications with Noise).

[0069] After completing the data association operation, the Kalman filter algorithm can be used to perform Kalman filtering on the associated point cloud data. Specifically, a reasonable mathematical model is established for the target motion in the scene, and a two-dimensional uniform linear motion model can be selected. For each frame of point cloud data, the Kalman filter recursively estimates the target state through the target motion model and the association between the point cloud data and the tracking trajectory, thereby adjusting the prediction value according to the associated point cloud data to obtain the optimal state estimation, and updating the target state based on the optimal state estimation result. The target state can be the trajectory position and center velocity of the tracking body and the point cloud set contained in the tracking body of the current frame. Among them, the point cloud set is a set of data corresponding to the tracking body in the point cloud data.

[0070] In this embodiment, after obtaining the point cloud set, unconstrained sitting-standing conversion estimation can be performed based on the point clouds in the point cloud set. Figure 4 As shown, this step includes several processes: adaptive height segmentation, high-confidence point cloud filtering, micro-Doppler time map decomposition, long-term accumulation of standing point clouds, standing direction vector solution, and sit-to-stand transition estimation. Specifically, in order to remove noise and avoid affecting the target motion direction judgment process, a high-confidence point cloud can be extracted based on the signal-to-noise ratio, that is, a high-confidence point cloud filtering process is performed. The specific execution method of this process is to first sort the energy of the point cloud tracking the large motion state and select the energy calculation average within a preset range. For example, the average of the minimum energy value and the middle energy value can be used as the energy calculation average; then, the signal-to-noise ratio of each point cloud is calculated based on the energy calculation average, and the point cloud with a signal-to-noise ratio greater than a preset signal-to-noise ratio threshold is selected as the high-confidence point cloud. For example, the preset signal-to-noise ratio threshold can be 1. The above point cloud set is updated based on the point cloud with a signal-to-noise ratio greater than the preset signal-to-noise ratio threshold.

[0071] In a specific implementation, a micro-Doppler array corresponding to each frame of point cloud data can be generated based on the determined target state and the corresponding point cloud set. Specifically, the distance and Doppler indexes can be reversed based on the distance and Doppler information corresponding to each point cloud in the point cloud set. Each point cloud corresponds to a point in the range-Doppler map, which can be used to reversely generate the range-Doppler map of the torso. The range-Doppler maps corresponding to each position are superimposed to obtain a single-frame micro-Doppler array for the entire frame. After accumulating and normalizing the micro-Doppler arrays corresponding to multiple frames of point cloud data, a target micro-Doppler time map can be obtained. The target micro-Doppler time map can be the first micro-Doppler time map corresponding to the entire target object.

[0072] To avoid overlapping or intersecting limb curves in micro-Doppler time maps, the human body can be considered a three-segment model. The first segment, from the head to the hips, shows the standing motion trajectory near the head moving horizontally forward and then vertically upward. The second segment, from the hips to the knees, shows the trajectory near the hips moving diagonally upward. The third segment, from the knees to the feet, shows the trajectory of the knees emulating a pendulum. Therefore, the point cloud in the point cloud collection needs to be segmented by height to provide an adaptive threshold for decomposing the micro-Doppler time map and determining the standing direction.

[0073] Optionally, based on each frame of point cloud data, a target micro-Doppler time map corresponding to the target object is generated, including: determining the point cloud height corresponding to the micro-motion point cloud based on each frame of point cloud data; determining the chest threshold value corresponding to the target object based on the point cloud height; and generating a first micro-Doppler time map corresponding to the torso of the target object, a second micro-Doppler time map corresponding to the head area, and a third micro-Doppler time map corresponding to the hip area based on the chest threshold value and the point cloud data.

[0074] It should be noted that a stable micro-motion point cloud is generated when sitting still. The chest cavity is the main radar reflection area of the human body, and the micro-motion caused by breathing will cause the micro-motion point cloud to be concentrated around the chest cavity. The point cloud height of the micro-motion points can be tracked by accumulating multiple frames and performing regional statistics to perform an adaptive point cloud height division process, dividing the human body into three-section models. For example, the micro-motion point cloud can be determined based on the distance information corresponding to each point cloud in the point cloud set. In addition, the division is performed every 5 cm, and the upper boundary with the most points falling into the area is regarded as the chest height. For example, if the height value falling into the range of 1.05-1.10m is the most, then the chest height is 1.10m, and 1.10 is used as the chest threshold value T1 for point cloud screening in micro-Doppler time map generation and direction judgment. At the same time, the tracking coordinates are relatively stable when sitting still, and the coordinates when sitting still are used as the starting coordinates for standing up. Since the first and second segments of the human body primarily move during the sit-to-stand transition, the point cloud can be segmented based on the thoracic threshold value T1. This allows the micro-Doppler array corresponding to point cloud data with a height exceeding the thoracic threshold value T1 to be separated from the point cloud set. A second micro-Doppler time map is generated from this micro-Doppler array. Furthermore, a micro-Doppler array corresponding to point cloud data with a height below or equal to the thoracic threshold value T1 is segmented from the point cloud set. This micro-Doppler array is then used to generate a third micro-Doppler time map. Alternatively, micro-Doppler time map decomposition can be performed directly based on the first micro-Doppler time map to generate the second and third micro-Doppler time maps.

[0075] In this embodiment, considering that the sit-to-stand transition process mainly relies on the movement of the part above the hips of the target object, a first micro-Doppler time map, a second micro-Doppler time map, and a third micro-Doppler time map are generated based on the chest position of the target object, respectively. This is conducive to analyzing the motion information of the target object from different areas and improving the accuracy of determining the sit-to-stand transition information.

[0076] S220 : Determine, based on the target micro-Doppler time map corresponding to each frame of the point cloud data, a micro-cluster point cloud dataset corresponding to a balance phase during the sit-to-stand transition movement of the target object.

[0077] It should be noted that the sit-to-stand transition process may include a balance phase, a rise phase, and a standing phase. The balance phase reflects the target object's body leaning forward before transitioning from sitting to standing, the rise phase reflects the target object's body leaning forward before transitioning to standing, and the holding phase reflects the target object's body standing. During the balance phase, the displacement of the upper torso is more obvious, the Doppler energy is higher, and there is a clearer large-motion point cloud. These features can be used to determine the direction of standing up later. Those skilled in the art will appreciate that the Doppler in the balance phase will always increase first and then decrease. Therefore, point clouds can be accumulated while generating micro-Doppler time maps to more accurately capture the balance phase process.

[0078] Specifically, since the micro-Doppler time map is obtained from point cloud data, there is no background noise in the micro-Doppler time map. Figure 5 A micro-Doppler time diagram is provided in an embodiment of the present invention, such as Figure 5 As shown, the envelope in the micro-Doppler time map includes an upper envelope and a lower envelope. By searching the first non-zero index and the last non-zero index of the micro-Doppler time map from top to bottom in each frame, the envelope feature can be quickly extracted.

[0079] The process of determining the micro-cluster point cloud dataset corresponding to the equilibrium stage is Figure 4 In the accumulation process of the standing point cloud duration described in the above, each frame on the second micro-Doppler time map is traversed. For the currently traversed frame, the upper and lower envelopes corresponding to the frame on the second micro-Doppler time map are compared with each other in terms of Doppler absolute value. The maximum value obtained by comparison is recorded as D m At the same time, the Doppler absolute value corresponding to the energy maximum value corresponding to the second micro-Doppler time map can be extracted. m D of the previous frame m Compare, if the D of the current frame m Greater than the previous frame D m , it means that the peak value has not been reached yet, and D is updated at this time. m , and accumulate the point cloud data of the large moving point cloud whose height is greater than the chest threshold value in the current frame and previous frames, and reset the counter at the same time. m Less than or equal to D of the previous frame m , indicating that the peak has been reached, point cloud data for the large dynamic point cloud can continue to be accumulated, and the counter will be incremented by 1. When the counter reaches the preset value, point cloud data collection and peak search can be stopped, and the peak time corresponding to the peak is recorded. The point cloud data accumulated before the peak time will form the micro-cluster point cloud dataset corresponding to the equilibrium stage. The preset value can be 5 or 10.

[0080] S230: Analyze the micro-cluster point cloud dataset based on a principal component analysis method, and determine the target movement direction of the target object when performing the sit-to-stand transition movement based on the analysis result.

[0081] In this embodiment, after determining that it is a micro-cluster point cloud dataset, the standing direction vector can be solved, that is, the target movement direction can be determined. Specifically, the principal components analysis (PCA) method is used to determine the direction of the micro-cluster point cloud dataset. PCA can be summarized as follows: the covariance matrix of the point cloud in the micro-cluster point cloud dataset in the xy plane is calculated, and the eigenvalues and eigenvectors of the covariance matrix are calculated. The eigenvalue reflects the variance of the point cloud in that direction, and the eigenvector represents the main distribution trend of the point cloud in that direction. By comparing the eigenvalues in different directions, the direction with the largest eigenvalue is selected as the main direction of the point cloud. This main direction is the target movement direction when the target object performs the sit-to-stand transition movement.

[0082] In this embodiment, by extracting the micro-cluster point cloud dataset corresponding to the balance stage, the characteristics of the sit-to-stand transition stage can be highlighted; and the principal component analysis method is used to analyze the micro-cluster point cloud dataset to effectively and quickly determine the target movement direction.

[0083] S240: Determine a target information extraction method corresponding to the target motion direction from among a plurality of pre-set information extraction methods.

[0084] S250: Determine the target object's sitting-standing conversion information based on the target information extraction method and the radar echo signal.

[0085] The sit-to-stand transition information is the speed information of the target object during the sit-to-stand transition process.

[0086] Optionally, the information extraction method includes a first extraction method corresponding to when the motion direction is facing the radar direction; determining the target object's sitting-standing transition information based on the target information extraction method and the radar echo signal includes: when the target information extraction method is the first extraction method, determining a first time corresponding to a maximum value of the envelope of the third micro-Doppler time map; determining a second time corresponding to a minimum value of the envelope of the third micro-Doppler time map based on the first time and a predetermined peak time; determining the target object's sitting-standing transition information before the second time based on the first micro-Doppler time map, and determining the target object's sitting-standing transition information after the second time based on the second micro-Doppler time map and the third micro-Doppler time map. The peak time is the acquisition time corresponding to the last frame of the point cloud in the micro-cluster point cloud dataset.

[0087] Specifically, the first moment is the moment after the peak moment. The second moment corresponding to the minimum envelope value of the third micro-Doppler time graph between the peak moment and the first moment is determined. When extracting features from the micro-Doppler time graph, the lower envelope of the first micro-Doppler time graph before the second moment is used as the head velocity curve of the target subject, and the hip velocity before the second moment is maintained at 0. After the second moment, the maximum value between the absolute value of the lower envelope and the absolute value of the upper envelope of the second micro-Doppler time graph after the second moment is determined. The head velocity curve after the second moment is composed of these maximum values corresponding to each moment after the second moment, and the lower envelope of the third micro-Doppler time graph after the second moment is used as the hip velocity curve after the second moment. Based on the screened head velocity curve and hip velocity curve, the sit-to-stand transition information of the target subject is determined.

[0088] This embodiment provides a method for determining sit-stand conversion information when facing radar movement, so as to quickly and accurately determine the sit-stand conversion information for the target object when facing radar movement.

[0089] Optionally, the information extraction method includes a second extraction method corresponding to when the movement direction is away from the radar direction; based on the target information extraction method and the radar echo signal, the sit-stand conversion information of the target object is determined, including: when the target information extraction method is the second extraction method, based on the predetermined peak moment and the first micro-Doppler time map, the third moment is determined; before the peak moment and after the third moment, the sit-stand conversion information is determined based on the first micro-Doppler time map; between the peak moment and the third moment, the sit-stand conversion information is determined based on the first micro-Doppler time map and the third micro-Doppler time map.

[0090] In this embodiment, for situations where the direction of motion is facing away from the radar, the time corresponding to the minimum difference between the upper and lower envelopes in the first micro-Doppler time map after the peak moment is used as the third moment. The upper envelope of the first micro-Doppler time map before the peak moment is extracted as the head velocity curve, and the hip velocity is set to zero. After the peak moment and before the third moment, the corresponding upper envelope in the first micro-Doppler time map continues to be used as the head velocity curve, and the upper envelope in the third micro-Doppler time map is used as the hip velocity curve. After the third moment, the lower envelope in the first micro-Doppler time map is extracted as the head velocity curve, and the upper envelope in the first micro-Doppler time map is extracted as the hip velocity curve. Based on the determined head and hip velocity curves, sit-to-stand transition information for situations where the direction of motion is facing away from the radar is determined.

[0091] This embodiment provides a method for determining sit-to-stand conversion information when the target object moves with its back to the radar, thereby quickly and accurately determining the sit-to-stand conversion information when the target object moves with its back to the radar.

[0092] Furthermore, the information extraction method includes a third extraction method corresponding to when the movement direction is the side-facing radar direction; based on the target information extraction method and the radar echo signal, the sitting-standing conversion information of the target object is determined, including: when the target information extraction method corresponds to the side-facing radar direction type, the sitting-standing conversion information is determined based on the maximum energy of each frame of point cloud data corresponding to the first micro-Doppler time diagram.

[0093] It should be noted that since the movement direction of the side-facing radar is very close to the radar tangent direction, the Doppler values of each part have a high degree of overlap, and most of the Doppler values are located near the low frequency and zero frequency. Therefore, the entire process of the four side-facing radars directly uses the Doppler absolute value corresponding to the maximum energy at each moment in the first micro-Doppler time diagram to determine the overall velocity curve.

[0094] If data is lost during the process, an interpolation algorithm, such as linear interpolation, is used to supplement the data. Finally, the speed curve is processed using a smoothing algorithm to obtain the final speed curve. For example, the smoothing algorithm may include five-point cubic smoothing, moving average, Savitzky-Golay filter, Gaussian filter, etc. to obtain the final speed curve.

[0095] This embodiment determines the overall velocity of the target object by using the maximum energy value when the Doppler values are repeated, thereby providing a convenient and effective method for determining the velocity.

[0096] In addition, the sit-to-stand conversion information obtained when facing the radar, and the sit-to-stand conversion information obtained when facing away from the radar, can be corrected according to the angle between the target movement direction and the measurement direction. Specifically, the correction formula is as follows:

[0097]

[0098] The sit-stand conversion information may be a speed curve, V2 represents the speed value in the speed curve after correction, V1 represents the speed value in the speed curve before correction, and θ represents the angle between the target motion direction and the measurement direction.

[0099] Compared to existing sit-to-stand transition estimation technologies, this technical solution requires no additional equipment and significantly increases the target subject's freedom of movement. Furthermore, it is insensitive to lighting, dust, smoke, and temperature, and can be used 24 / 7, ensuring reliability and applicability in a variety of real-world scenarios. During point cloud data processing, Doppler effect variations are preserved, providing multi-dimensional feature information for applications such as posture estimation, motion classification, and identity recognition. Furthermore, the extracted velocity curves play an important role in home health monitoring and intelligent identification. By analyzing these curves, motion classification can be achieved, helping to distinguish between different activity states of the target subject, such as walking, sitting, standing, or falling. It can also be used for identity recognition, such as using micro-Doppler time curves between different individuals for machine learning or deep learning training to determine individual identity. Velocity curves help assess a subject's motor ability and activity level. Through continuous monitoring, changes in the target subject's movement patterns can be identified, allowing for timely detection of unstable standing or sudden acceleration changes, thereby predicting fall risk and enabling healthcare professionals to adjust treatment or care plans in real time, improving home safety and quality of life, thereby enhancing overall home health management effectiveness.

[0100] Figure 6 This is a schematic diagram of the structure of a device for determining sit-to-stand transition information provided in accordance with an embodiment of the present invention. The device is used to execute the sit-to-stand transition information determination method provided in any of the above embodiments. The device and the sit-to-stand transition information determination method of the above embodiments belong to the same inventive concept. For details not fully described in the embodiments of the sit-to-stand transition information determination device, please refer to the embodiments of the sit-to-stand transition information determination method. Figure 6 As shown, the device includes:

[0101] a motion direction determination module 10 for determining the target motion direction of the target object when performing a sit-to-stand transition motion based on a received radar echo signal; wherein the radar echo signal is measurement data obtained by performing a radar measurement of the target object's environment;

[0102] An extraction mode determination module 11 is used to determine a target information extraction mode corresponding to the target motion direction from among a plurality of pre-set information extraction modes;

[0103] The information determination module 12 is used to determine the sitting-standing conversion information of the target object based on the target information extraction method and the radar echo signal.

[0104] Based on any optional technical solution in the embodiment of the present invention, optionally, the motion direction determination module 10 includes:

[0105] The data generation submodule is used to generate multi-frame point cloud data corresponding to the target object based on the radar echo signal;

[0106] The time map generation submodule is used to generate a target micro-Doppler time map corresponding to the target object based on each frame of point cloud data;

[0107] The dataset determination submodule is used to determine the micro-cluster point cloud dataset corresponding to the balance phase of the target object during the sit-to-stand transition based on the target micro-Doppler time map corresponding to each frame of point cloud data. The balance phase reflects the forward leaning phase of the target object before transitioning from sit-to-stand.

[0108] The motion direction determination submodule is used to analyze the micro-cluster point cloud dataset based on the principal component analysis method, and determine the target motion direction when the target object performs the sit-to-stand conversion motion based on the analysis results.

[0109] Based on any optional technical solution in the embodiments of the present invention, optionally, the data generation submodule includes:

[0110] A first signal processing unit is used to perform a fast Fourier transform on each frame of the radar echo signal to obtain a first processed signal;

[0111] a signal accumulation unit configured to, for each frame of the first processed signal, use the first processed signal as a start signal, accumulate the start signal and a preset number of frames of the first processed signal after the start signal, and obtain a first signal group;

[0112] a second signal processing unit, configured to perform clutter suppression processing on each frame of the first processed signal, update the first processed signal based on a processing result corresponding to the first processed signal, perform clutter suppression processing on each first signal group, and update the first signal group based on a processing result corresponding to the first signal group;

[0113] a third signal processing unit, configured to obtain, based on the updated first processed signal, the updated first signal group, and a slow-time fast Fourier transform, a second processed signal corresponding to the first processed signal and a second signal group corresponding to the first signal group;

[0114] a target detection unit, configured to perform target detection on each second processed signal and a second signal group corresponding to the second processed signal based on a constant false alarm algorithm, to obtain first detection data corresponding to the second processed signal and second detection data corresponding to the second signal group;

[0115] An angle estimation unit, configured to perform two-dimensional angle estimation processing on the first detection data and the second detection data, respectively, to obtain angle data corresponding to the first detection data and angle data corresponding to the second detection data;

[0116] a data generating unit configured to, for each first detection data, form first point cloud data from the first detection data and angle data corresponding to the first detection data; and, for each second detection data, form second point cloud data from the second detection data and angle data corresponding to the second detection data;

[0117] The first detection data and the second detection data respectively include the distance between the target object and the radar, the Doppler data of the target object relative to the radar, and the signal energy data; the angle data includes the azimuth angle and the pitch angle.

[0118] Based on any optional technical solution in the embodiments of the present invention, optionally, the time graph generation submodule includes:

[0119] A height determination unit, configured to determine the point cloud height corresponding to the micro-motion point cloud based on each frame of point cloud data;

[0120] A chest threshold value determination unit, configured to determine a chest threshold value corresponding to the target object based on the point cloud height;

[0121] The time map generating unit is used to generate a first micro-Doppler time map corresponding to the torso, a second micro-Doppler time map corresponding to the head area, and a third micro-Doppler time map corresponding to the buttocks area of the target object based on the chest threshold value and the point cloud data.

[0122] Based on any optional technical solution in the embodiment of the present invention, optionally, the information extraction method includes a first extraction method corresponding to when the motion direction is facing the radar direction;

[0123] The information determination module 12 includes:

[0124] A first moment determination submodule, configured to determine a first moment corresponding to an envelope maximum value of the third micro-Doppler time map when the target information extraction mode is the first extraction mode;

[0125] a second time determination submodule, configured to determine a second time corresponding to an envelope minimum value of the third micro-Doppler time map based on the first time and a predetermined peak time;

[0126] a first information determination submodule, configured to determine the target object's sitting-standing transition information before a second moment based on the first micro-Doppler time map, and determine the target object's sitting-standing transition information after the second moment based on the second micro-Doppler time map and the third micro-Doppler time map;

[0127] Among them, the peak moment is the acquisition moment corresponding to the last frame of point cloud in the micro-cluster point cloud dataset.

[0128] On the basis of any optional technical solution in the embodiment of the present invention, optionally, the information extraction method includes a second extraction method corresponding to when the motion direction is facing away from the radar direction;

[0129] The information determination module 12 includes:

[0130] a third time determination submodule, configured to determine a third time based on a predetermined peak time and the first micro-Doppler time map when the target information extraction mode is the second extraction mode;

[0131] a second information determination submodule, configured to determine sit-to-stand transition information based on the first micro-Doppler time map before the peak moment and after the third moment;

[0132] a third information determination submodule, configured to determine, between the peak moment and the third moment, sit-to-stand transition information based on the first micro-Doppler time map and the third micro-Doppler time map;

[0133] Among them, the peak moment is the acquisition moment corresponding to the last frame of point cloud in the micro-cluster point cloud dataset.

[0134] On the basis of any optional technical solution in the embodiment of the present invention, optionally, the information extraction method includes a third extraction method corresponding to when the motion direction is sideways to the radar direction;

[0135] The information determination module 12 includes:

[0136] The fourth information determination submodule is used to determine the sitting-standing conversion information based on the maximum energy of each frame of point cloud data corresponding to the first micro-Doppler time map when the target information raising method corresponds to the side radar direction type.

[0137] Based on any optional technical solution in the embodiment of the present invention, optionally, the extraction method determination module 11 includes:

[0138] A measurement direction determination submodule is used to determine the direction of the line corresponding to the connection between the target object and the radar based on the radar echo signal as the measurement direction;

[0139] a target direction type determination submodule, configured to determine an angle between a target motion direction and a measurement direction, and determine a target direction type of the target motion direction based on the angle;

[0140] An extraction method determination submodule, configured to use the information extraction method corresponding to the target direction type as the target information extraction method based on a predetermined correspondence between the direction type and the information extraction method;

[0141] The target direction type includes at least one of a facing radar direction type, a back-facing radar direction type, and a side-facing radar direction type.

[0142] The technical solution of an embodiment of the present invention determines the target motion direction of a target object when performing a sit-to-stand transition based on a received radar echo signal. The radar echo signal is measurement data obtained by performing radar measurement of the target object's environment. Furthermore, a target information extraction method corresponding to the target motion direction is determined from a plurality of pre-set information extraction methods. This allows for flexible determination of the corresponding information extraction method based on the target object's motion direction, without requiring the target object to move in a pre-set motion direction. Furthermore, the target object's sit-to-stand transition information is determined based on the target information extraction method and the radar echo signal. The technical solution of this embodiment eliminates the need to constrain the target object's motion direction, accommodates the diverse motion patterns of the target object in real life, and can effectively and accurately determine the target object's sit-to-stand transition information in daily life.

[0143] It is worth noting that in the embodiment of the above-mentioned sit-stand conversion information determination device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the various functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0144] Figure 7 1 is a schematic diagram of the structure of an electronic device that implements the sit-stand transition information determination method of an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0145] like Figure 7As shown, the electronic device 20 includes at least one processor 21, and a memory connected to the at least one processor 21, such as a read-only memory (ROM) 22, a random access memory (RAM) 23, etc., wherein the memory stores a computer program that can be executed by the at least one processor, and the processor 21 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 22 or the computer program loaded from the storage unit 28 to the random access memory (RAM) 23. Various programs and data required for the operation of the electronic device 20 can also be stored in the RAM 23. The processor 21, ROM 22 and RAM 23 are connected to each other via a bus 24. An input / output (I / O) interface 25 is also connected to the bus 24.

[0146] Multiple components in the electronic device 20 are connected to the I / O interface 25, including an input unit 26, such as a keyboard, a mouse, etc.; an output unit 27, such as various types of displays, speakers, etc.; a storage unit 28, such as a magnetic disk, an optical disk, etc.; and a communication unit 29, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 29 allows the electronic device 20 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0147] The processor 21 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 21 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors that run machine learning model algorithms, a digital signal processor (DSP), and any other suitable processor, controller, microcontroller, etc. The processor 21 executes the various methods and processes described above, such as the method for determining sit-to-stand transition information.

[0148] In some embodiments, the sit-to-stand transition information determination method can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 28. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 20 via the ROM 22 and / or the communication unit 29. When the computer program is loaded into the RAM 23 and executed by the processor 21, one or more steps of the sit-to-stand transition information determination method described above can be performed. Alternatively, in other embodiments, the processor 21 can be configured to perform the sit-to-stand transition information determination method in any other suitable manner (e.g., via firmware).

[0149] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips or systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0150] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0151] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0152] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0153] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0154] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0155] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product that includes a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network via the communication unit 29, or installed from the storage unit 28, or installed from the ROM 22. When the computer program is executed by the processor 21, the above-mentioned functions defined in the method of the embodiment of the present invention are performed.

[0156] The computer program product may be implemented by writing computer program code for performing the operations of the present invention in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0157] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0158] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A method for determining sit-stand conversion information, characterized in that: include: Determining a target movement direction of a target object when performing a sit-to-stand transition movement based on a received radar echo signal, wherein the radar echo signal is measurement data obtained by performing radar measurement of an environment in which the target object is located; Determining a target information extraction method corresponding to the target motion direction from among a plurality of pre-set information extraction methods; Based on the target information extraction method and the radar echo signal, sit-stand conversion information of the target object is determined.

2. The method according to claim 1, characterized in that The determining, based on the received radar echo signal, the target movement direction of the target object when performing the sit-to-stand transition movement includes: generating a plurality of frames of point cloud data corresponding to the target object based on the radar echo signal; generating a target micro-Doppler time map corresponding to the target object based on each frame of the point cloud data; Based on the target micro-Doppler time map corresponding to each frame of the point cloud data, a micro-cluster point cloud dataset corresponding to a balance phase of the target object during the sit-to-stand transition movement is determined; wherein the balance phase reflects a forward leaning phase of the target object before transitioning from sitting to standing; The micro-cluster point cloud dataset is analyzed based on a principal component analysis method, and a target movement direction of the target object when performing the sit-to-stand transition movement is determined based on the analysis result.

3. The method according to claim 2, characterized in that The step of generating multi-frame point cloud data corresponding to the target object based on the radar echo signal includes: Performing a fast Fourier transform (FFT) on each frame of the radar echo signal to obtain a first processed signal. For each frame of the first processed signal, taking the first processed signal as a start signal, accumulating the start signal and a preset number of frames of first processed signals after the start signal to obtain a first signal group; performing clutter suppression processing on each frame of the first processed signal, updating the first processed signal based on a processing result corresponding to the first processed signal, performing clutter suppression processing on each first signal group, and updating the first signal group based on a processing result corresponding to the first signal group; Obtaining a second processed signal corresponding to the first processed signal and a second signal group corresponding to the first signal group based on the updated first processed signal, the updated first signal group, and a slow-time fast Fourier transform; For each second processed signal and the second signal group corresponding to the second processed signal, performing target detection on the second processed signal and the second signal group based on a constant false alarm algorithm to obtain first detection data corresponding to the second processed signal and second detection data corresponding to the second signal group; performing two-dimensional angle estimation processing on the first detection data and the second detection data respectively to obtain angle data corresponding to the first detection data and angle data corresponding to the second detection data; For each first detection data, forming first point cloud data by using the first detection data and angle data corresponding to the first detection data; For each piece of the second detection data, forming second point cloud data from the second detection data and the angle data corresponding to the second detection data; The first detection data and the second detection data respectively include the distance between the target object and the radar, the Doppler data of the target object relative to the radar, and the signal energy data; the angle data includes the azimuth angle and the pitch angle.

4. The method according to claim 2, characterized in that Generating a target micro-Doppler time map corresponding to the target object based on each frame of the point cloud data includes: Determining the point cloud height corresponding to the micro-motion point cloud based on the point cloud data of each frame; Determining a chest threshold value corresponding to the target object based on the point cloud height; Based on the chest threshold value and the point cloud data, a first micro-Doppler time map corresponding to the torso, a second micro-Doppler time map corresponding to the head region, and a third micro-Doppler time map corresponding to the buttocks region of the target object are generated.

5. The method according to claim 4, characterized in that The information extraction method includes a first extraction method corresponding to when the movement direction is facing the radar direction; The determining, based on the target information extraction method and the radar echo signal, the sit-stand conversion information of the target object includes: When the target information extraction mode is the first extraction mode, determining a first time corresponding to an envelope maximum value of the third micro-Doppler time map; Determining a second time corresponding to an envelope minimum value of the third micro-Doppler time map based on the first time and a predetermined peak time; determining the sitting-to-standing transition information of the target object before the second moment based on the first micro-Doppler time map, and determining the sitting-to-standing transition information of the target object after the second moment based on the second micro-Doppler time map and the third micro-Doppler time map; The peak moment is the acquisition moment corresponding to the last frame of point cloud in the micro-cluster point cloud dataset.

6. The method according to claim 4, characterized in that The information extraction method includes a second extraction method corresponding to when the motion direction is facing away from the radar direction; The determining, based on the target information extraction method and the radar echo signal, the sit-stand conversion information of the target object includes: When the target information extraction method is the second extraction method, determining a third time based on a predetermined peak time and the first micro-Doppler time map; determining the sit-to-stand transition information based on the first micro-Doppler time map before the peak moment and after the third moment; between the peak moment and the third moment, determining the sit-to-stand transition information based on the first micro-Doppler time map and the third micro-Doppler time map; The peak moment is the acquisition moment corresponding to the last frame of point cloud in the micro-cluster point cloud dataset.

7. The method according to claim 4, characterized in that The information extraction method includes a third extraction method corresponding to when the movement direction is sideways to the radar direction; The determining, based on the target information extraction method and the radar echo signal, the sit-stand conversion information of the target object includes: In a case where the target information lifting method corresponds to the side-facing radar direction type, the sitting-to-standing conversion information is determined based on the maximum energy value of each frame of the point cloud data corresponding to the first micro-Doppler time map.

8. The method according to claim 1, characterized in that Determining the target information extraction method corresponding to the target motion direction from among the pre-set multiple information extraction methods includes: Based on the radar echo signal, determining a connection direction corresponding to the connection between the target object and the radar as a measurement direction; determining an angle between the target motion direction and the measurement direction, and determining a target direction type of the target motion direction based on the angle; Based on a preset correspondence between direction types and information extraction methods, using the information extraction method corresponding to the target direction type as the target information extraction method; The target direction type includes at least one of a facing radar direction type, a back-facing radar direction type, and a side-facing radar direction type.

9. A device for determining sit-stand transition information, characterized in that: include: a motion direction determination module, configured to determine the target motion direction of the target object when performing a sit-to-stand transition motion based on a received radar echo signal; wherein the radar echo signal is measurement data obtained by performing radar measurement of the target object's environment; An extraction mode determination module, configured to determine a target information extraction mode corresponding to the target motion direction from among a plurality of pre-set information extraction modes; An information determination module is used to determine the sitting-standing conversion information of the target object based on the target information extraction method and the radar echo signal.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the sit-to-stand transition information determination method according to any one of claims 1 to 8.

Images