Human body part recognition method, mattress and storage medium

Abstract

The embodiment of the invention discloses a human body part recognition method, a mattress and a storage medium, and the method comprises the steps: obtaining the three-dimensional contour data of a human body on the mattress through a laser radar sensor, and extracting the two-dimensional projection coordinates of a human skeleton according to the three-dimensional contour data; the whole pressure distribution data of the mattress is obtained through a piezoelectric film sensor, multiple target areas are divided according to the whole pressure distribution data and the two-dimensional projection coordinates of the human skeleton, and the multiple target areas comprise a head area, a trunk area and a limb area; obtaining local pressure distribution data of a local area of each target area through an air pressure sensor, and obtaining a central point of each target area according to the local pressure distribution data; and according to the two-dimensional projection coordinates of the human skeleton and the central point of each target area, reconstructing each body part of the human body, and identifying the position of each body part of the human body on the mattress. By implementing the embodiment of the invention, the position of each part of the human body on the mattress can be accurately identified.

Description

Technical Field

[0001] The present application relates to the field of intelligent identification technology, and is related to, but not limited to, a method for identifying human body parts, a mattress, and a storage medium. Background Art

[0002] As demand for healthy sleep and smart furniture grows, identifying the position of various body parts on mattresses has become a key technology for improving the sleeping experience. During sleep, due to the body's physiological structure and mechanical properties, different parts of the body require different support from the mattress. Therefore, to provide users with the best sleeping experience, it is necessary to accurately identify these key parts of the body and provide the appropriate support level for each part.

[0003] In related technologies, pressure sensors can be used to detect the pressure distribution of various parts of the human body on the mattress, thereby providing position information of various parts of the human body on the mattress to a certain extent. However, the resolution of pressure sensors is usually low, and they cannot accurately distinguish the pressure boundaries of adjacent parts, resulting in inaccurate identification results of the positions of various parts of the human body on the mattress. Therefore, how to accurately identify the positions of various parts of the human body on the mattress is a technical problem that needs to be solved urgently. Summary of the Invention

[0004] In view of this, the embodiments of the present application provide a human body part recognition method, a mattress, and a storage medium, which can accurately identify the positions of various human body parts on the mattress.

[0005] In a first aspect, an embodiment of the present application discloses a method for identifying a human body part, which is applied to a mattress. The mattress includes a piezoelectric film sensor and an air pressure sensor. A laser radar sensor is disposed above the mattress and is in communication with the mattress. The method includes:

[0006] Acquiring three-dimensional contour data of a human body on the mattress by the laser radar sensor, and extracting two-dimensional projection coordinates of the human skeleton based on the three-dimensional contour data;

[0007] Acquiring overall pressure distribution data of the mattress through the piezoelectric film sensor, and dividing the mattress into multiple target areas according to the overall pressure distribution data and the two-dimensional projection coordinates of the human skeleton, wherein the multiple target areas include a head area, a torso area, and a limb area;

[0008] Acquire local pressure distribution data of a local area of each target area through the air pressure sensor, and obtain the center point of each target area according to the local pressure distribution data, wherein the local area of each target area is an area in the corresponding target area where the pressure data is greater than or equal to a threshold value, and the center point of each target area is a position corresponding to the maximum value of the pressure data in the corresponding local area;

[0009] According to the two-dimensional projection coordinates of the human skeleton and the center point of each target area, the various body parts of the human body are reconstructed, and the positions of the various body parts of the human body on the mattress are identified.

[0010] In some possible embodiments, the mattress includes multiple airbag units, the air pressure sensor includes multiple air pressure sensor submodules, each air pressure sensor submodule is used to monitor the air pressure value of at least one airbag unit, and obtaining local pressure distribution data of a local area of each target area through the air pressure sensor includes:

[0011] Increase the number of connections of the activated target air pressure sensor submodules in the local area of each target area, or dynamically connect the air pressure sensor submodule to the airbag unit of the local area through an air path switching device to adjust the monitoring position of the air pressure sensor submodule to obtain the local pressure distribution data.

[0012] In some possible embodiments, the air pressure sensor is disposed inside the airbag unit, and increasing the number of connected target air pressure sensor submodules in an activated state in a local area of each target area includes:

[0013] Calculating a target number of air pressure sensor submodules to be added based on a preset sensor density threshold, the area of the local area, and the number of currently activated air pressure sensor submodules in the local area;

[0014] A target number of air pressure sensor submodules in a dormant state in the local area are activated so that the number of connection of the target air pressure sensor submodules in the local area reaches the preset sensor density threshold.

[0015] In some possible embodiments, the mattress includes multiple air tubes, each of which is connected to the airbag unit. The air path switching device includes multiple micro air valves, each of which is used to control the passage of the air tube. The air pressure sensor submodule is dynamically connected to the airbag unit in the local area through the air path switching device to adjust the monitoring position of the air pressure sensor submodule, including:

[0016] determining a target airbag unit corresponding to a local area of each target area;

[0017] The micro air valve corresponding to the target airbag unit is controlled to be in an open state, so that the air pressure sensor submodule is connected to the target airbag unit through the air pipe, so as to adjust the monitoring position of the air pressure sensor submodule.

[0018] In some possible embodiments, dividing the target areas according to the overall pressure distribution data and the two-dimensional projection coordinates of the human skeleton includes:

[0019] Gridding the overall pressure distribution data to divide it into multiple pressure sub-regions, and calculating the pressure value and area of each pressure sub-region;

[0020] Calculating the pressure center coordinates of the overall pressure distribution data using a weighted average algorithm based on the pressure value and area of each pressure sub-region;

[0021] When the pressure difference between adjacent pressure sub-areas is less than the pressure gradient threshold, the adjacent pressure sub-areas are divided into the same pressure area, and the multiple target areas are divided according to the pressure center coordinates, the two-dimensional projection coordinates of the human skeleton, the position, area and pressure value distribution characteristics of the same pressure area.

[0022] In some possible embodiments, reconstructing each body part of the human body based on the two-dimensional projection coordinates of the human skeleton and the center point of each target area, and identifying the position of each body part of the human body on the mattress includes:

[0023] Mapping the position of the center point of each target area on the mattress to the position of the key points of the human skeleton;

[0024] According to the two-dimensional projection coordinates of the human skeleton and the positions of the key points of the skeleton, various body parts of the human body are reconstructed to obtain the positions of the various body parts of the human body on the mattress.

[0025] In some possible embodiments, extracting the two-dimensional projection coordinates of the human skeleton based on the three-dimensional contour data includes:

[0026] Mapping each data point in the three-dimensional contour data to the mattress plane coordinate system through projection transformation to obtain corresponding two-dimensional projection data;

[0027] The two-dimensional projection data is input into a pre-trained human posture estimation model, the human skeleton is extracted, and the two-dimensional projection coordinates of the human skeleton are obtained. The human posture estimation model is obtained by training an initial model based on the human posture and the human skeleton corresponding to the human posture.

[0028] A second aspect of an embodiment of the present application discloses a mattress, the mattress including a piezoelectric film sensor and an air pressure sensor, a laser radar sensor disposed above the mattress, the laser radar sensor being communicatively connected to the mattress, and the mattress including:

[0029] a first processing module, configured to obtain three-dimensional contour data of a human body on the mattress through the laser radar sensor, and extract two-dimensional projection coordinates of human bones based on the three-dimensional contour data;

[0030] a second processing module, configured to obtain overall pressure distribution data of the mattress through the piezoelectric film sensor, and divide the mattress into a plurality of target areas according to the overall pressure distribution data and the two-dimensional projection coordinates of the human skeleton, the plurality of target areas including a head area, a torso area, and a limb area;

[0031] a third processing module, configured to obtain local pressure distribution data of a local area of each target area through the air pressure sensor, and obtain a center point of each target area based on the local pressure distribution data, wherein the local area of each target area is an area in the corresponding target area where the pressure data is greater than or equal to a threshold, and the center point of each target area is a position corresponding to the maximum value of the pressure data in the corresponding local area;

[0032] The recognition module is used to reconstruct various body parts of the human body according to the two-dimensional projection coordinates of the human skeleton and the center point of each target area, and to identify the position of each body part of the human body on the mattress.

[0033] A third aspect of an embodiment of the present application discloses a mattress, which includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor implements the method described above.

[0034] A fourth aspect of an embodiment of the present application discloses a computer-readable storage medium having a computer program stored thereon, which implements the method described above when the computer program is executed by a processor.

[0035] Compared with the related art, the embodiments of the present application have at least the following beneficial effects:

[0036] The three-dimensional contour data of the human body on the mattress is obtained through a lidar sensor, and the two-dimensional projection coordinates of the human skeleton are extracted based on the three-dimensional contour data. The overall pressure distribution data of the mattress is obtained through a piezoelectric film sensor. Multiple target areas are divided based on the overall pressure distribution data and the two-dimensional projection coordinates of the human skeleton. The local pressure distribution data of a local area of each target area is obtained through an air pressure sensor. The center point of each target area is obtained based on the local pressure distribution data. The various body parts of the human body are reconstructed based on the two-dimensional projection coordinates of the human skeleton and the center point of each target area, and the positions of the various body parts of the human body on the mattress are identified.

[0037] In an embodiment of the present application, the two-dimensional projection coordinates of the human skeleton are extracted from the three-dimensional contour data obtained by the lidar sensor, providing overall structural information for human body part identification. The overall pressure distribution data of the piezoelectric film sensor is combined with the bone projection coordinates to divide target areas such as the head, torso and limbs. The air pressure sensor further obtains the local pressure distribution data of each target area, determines the center point of each area, and can accurately locate the key parts of the human body. Through multi-dimensional data fusion and analysis, the position of each part of the human body on the mattress can be accurately identified. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The drawings herein are incorporated into and constitute a part of the specification. These drawings illustrate embodiments consistent with the present application and, together with the specification, are used to illustrate the technical solutions of the present application.

[0039] Figure 1 A diagram illustrating an application scenario of the human body part recognition method provided in an embodiment of the present application;

[0040] Figure 2 A schematic diagram of a flow chart of a method for identifying human body parts provided in an embodiment of the present application;

[0041] Figure 3 A flowchart of a method for dividing multiple target areas provided in an embodiment of the present application;

[0042] Figure 4 A schematic diagram of dividing multiple target areas provided in an embodiment of the present application;

[0043] Figure 5 A schematic flow chart of obtaining local pressure distribution data provided in an embodiment of the present application;

[0044] Figure 6 A schematic diagram of the structure of a mattress provided in an embodiment of the present application;

[0045] Figure 7 This is a structural block diagram of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the specific technical solutions of the present application will be further described in detail below in conjunction with the drawings in the embodiments of the present application. The following embodiments are used to illustrate the present application but are not intended to limit the scope of the present application.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0048] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0049] It should be pointed out that the terms "first\second\third" involved in the embodiments of the present application are used to distinguish similar or different objects, and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described here can be implemented in an order other than that illustrated or described here.

[0050] The present invention provides a method for identifying human body parts, a mattress, and a storage medium, which can accurately identify the positions of various human body parts on the mattress. Detailed descriptions are provided below.

[0051] See also Figure 1 , Figure 1 This is an application scenario diagram of the human body part recognition method provided in the embodiment of the present application, such as Figure 1As shown, the mattress 10 includes a piezoelectric film sensor 101 and an air pressure sensor 102. A lidar sensor 20 is positioned above the mattress 10 and is in communication with the mattress 10. Optionally, the piezoelectric film sensor 101 can be installed inside or on the surface of the mattress 10, typically within the support layer or inside the mattress cover, to ensure comprehensive monitoring of pressure distribution data on the mattress 10. In some embodiments, the mattress is provided with multiple airbag units, which can be grouped to correspond to different parts of the human body. It is understood that the air pressure sensor 102 can obtain pressure distribution data on the human body by monitoring the air pressure in the airbag units. The higher the air pressure, the greater the pressure data. Optionally, a control box is provided at the foot of the bed. The control box integrates an air pump, a solenoid valve, and a circuit board. The control box is connected to the multiple airbag units via air pipes to control inflation and deflation. The air pressure sensor 102 can be installed on the circuit board and connected to the airbag units via air pipes to monitor the internal air pressure of the airbag units. The air pressure sensor 102 can also be located inside the airbag units. The lidar sensor 20 can be installed on the ceiling or bracket directly above the mattress 10 to ensure that the scanning range can cover the entire mattress surface, so as to ensure that the three-dimensional contour data of the human body on the mattress 10 can be fully obtained.

[0052] In some embodiments, the lidar sensor 20 is communicatively connected to the mattress 10. Exemplarily, the mattress 10 further includes a microprocessor and a communication module. The microprocessor can process and analyze the data collected by each sensor to identify the position of each body part on the mattress 10. The mattress 10 can establish a communication connection with the lidar sensor 20 through the communication module. The communication module may include but is not limited to a Bluetooth Low Energy (BLE) module, a Wireless Fidelity (Wi-Fi) module, a cellular module, a Low Power Wide Area Network (LPWAN) module, a ZigBee module, etc., wherein the cellular module may include but is not limited to the 4th Generation Mobile Communication Technology (4G) and the 5th Generation Mobile Communication Technology (5G).

[0053] See also Figure 2 , Figure 2 The flowchart of a method for identifying human body parts provided in the embodiment of the present application can accurately identify the position of various parts of the human body on the mattress, such as Figure 2As shown, the method may include the following steps:

[0054] Step 201: Acquire three-dimensional contour data of a human body on a mattress through a laser radar sensor, and extract the two-dimensional projection coordinates of the human skeleton based on the three-dimensional contour data.

[0055] In the embodiments of the present application, a lidar sensor emits a laser beam and measures the return time of the reflected light. By recording the time difference between laser beam emission and reception and incorporating the principle of the constant speed of light, the distance from the lidar sensor to each point on the human body surface can be calculated. The lidar repeatedly transmits and receives at a specific angle and frequency, acquiring a large number of distance data points. These data points form a three-dimensional point cloud of the human body on the mattress. This 3D point cloud data contains depth information and spatial position information of the human surface, forming a three-dimensional contour data of the human body on the mattress. The lidar sensor can transmit this 3D contour data to the mattress. It can be understood that this 3D contour data can intuitively present information such as the human body's appearance and posture.

[0056] In some embodiments, extracting the two-dimensional projection coordinates of the human skeleton based on the three-dimensional contour data includes:

[0057] Mapping each data point in the three-dimensional contour data to the mattress plane coordinate system through projection transformation to obtain corresponding two-dimensional projection data;

[0058] The two-dimensional projection data is input into the pre-trained human posture estimation model, the human skeleton is extracted, and the two-dimensional projection coordinates of the human skeleton are obtained. The human posture estimation model is obtained by training the initial model based on the human posture and the human skeleton corresponding to the human posture.

[0059] In the embodiment of the present application, each point cloud data point (x, y, z) in the three-dimensional contour data can be mapped to the mattress plane coordinate system through a projection transformation. Optionally, the mattress plane coordinate system takes the center point of the mattress as the origin, and the x-axis and y-axis are parallel to the length and width of the mattress, respectively. The projection transformation adopts an orthogonal projection method, and the specific calculation method is as follows:

[0060] First, ignore the height information of the point cloud data points and directly map (x, y, z) to two-dimensional coordinates (x′, y′). Second, consider the relative position relationship between the lidar and the mattress plane, coordinate translation and scaling are required. The calculation formula can be expressed as follows:

[0061]

[0062] Among them, (x offset ,y offset ) represents the offset of the origin of the laser radar coordinate system in the mattress plane coordinate system, scale x、scale y is the scaling factor, which is used to adjust the proportional relationship between the projected two-dimensional data and the actual size of the mattress. After the projection transformation, the two-dimensional projection data (x′, y′) of the three-dimensional contour of the human body on the mattress plane are obtained. These data points constitute the two-dimensional contour image of the human body.

[0063] In some embodiments, the constructed human posture estimation model can be trained on an initial model based on a large amount of labeled human postures and corresponding skeletal data, so that the model can learn the mapping relationship between human posture and skeletal position under different body shapes and postures. Optionally, human postures can include multiple sleeping positions, such as supine, side-lying, and prone human postures. By inputting the two-dimensional projection data of the human body into the pre-trained human posture estimation model, the model automatically extracts features through an internal multi-layer neural network and outputs the two-dimensional projection coordinates of the human skeleton in the mattress plane coordinate system. Exemplarily, the initial model can be a convolutional neural network model, a Transformer network, a graph neural network, etc.

[0064] In the above embodiment, a lidar sensor acquires real-time 3D contour data of the human body on the mattress and converts this 3D contour data into 2D projection data. This allows for intuitive and accurate visualization of the human body's shape, based on the mattress's plane coordinate system. A pre-trained human pose estimation model can adaptively handle complex poses and body shape differences, accurately identifying the human skeleton, thereby improving the accuracy and real-time performance of identifying the positions of various body parts on the mattress.

[0065] In step 202, the overall pressure distribution data of the mattress is obtained through a piezoelectric film sensor, and multiple target areas are divided according to the overall pressure distribution data and the two-dimensional projection coordinates of the human skeleton. The multiple target areas include a head area, a torso area, and a limb area.

[0066] In some embodiments, the piezoelectric film sensor array can be evenly arranged in a grid on the surface of the mattress to collect real-time pressure distribution data of the user on the mattress. For example, the piezoelectric film sensor can convert the pressure data into an electrical signal, thereby generating a two-dimensional pressure distribution map, which reflects the overall pressure distribution of the human body on the mattress.

[0067] In some embodiments, a flow chart of a method for dividing multiple target areas according to the overall pressure distribution data and the two-dimensional projection coordinates of the human skeleton is shown in FIG. Figure 3 As shown, the multiple target areas include a head area, a torso area, and a limb area. Figure 3 The method shown may include the following steps:

[0068] Step 301 : gridding the overall pressure distribution data, dividing it into multiple pressure sub-regions, and calculating the pressure value and area of each pressure sub-region.

[0069] In an embodiment of the present application, the overall pressure distribution data on the mattress can be divided into multiple regular grid cells according to the spatial dimension. Each grid cell corresponds to a pressure sub-region. The size of the grid cell can be set according to actual needs, for example, 5 cm × 5 cm, 10 cm × 10 cm, etc. For each pressure sub-region, the electrical signal collected by the piezoelectric film sensor is converted into a pressure value. If a pressure sub-region contains multiple piezoelectric film sensors, the average of the pressure values collected by these sensors can be taken as the pressure value for that sub-region. It can be understood that the area of each grid cell is the area of each pressure sub-region.

[0070] Step 302 : Calculate the pressure center coordinates of the overall pressure distribution data using a weighted average algorithm based on the pressure value and area of each pressure sub-region.

[0071] As an optional implementation, the pressure center coordinates are calculated using a weighted average algorithm, which can be calculated using the following formula:

[0072]

[0073]

[0074] Where i is the pressure sub-region index, n is the total number of pressure sub-regions, (x i ,y i ) represents the center coordinate of the i-th pressure sub-region, P i represents the pressure value of the i-th pressure sub-region, S i represents the area of the ith pressure sub-region, (x c ,y c ) represents the pressure center coordinates of the overall pressure distribution data, and

[0075] Represents the sum of the product of the pressure value and area of all pressure sub-areas on the x-axis and y-axis respectively, The total pressure is the sum of the product of the pressure values and the areas of all pressure sub-regions. By weighted averaging, sub-regions with larger pressure values have a greater impact on the pressure center coordinate, ensuring that the pressure center coordinate of the overall pressure distribution data accurately reflects the center of gravity of the overall pressure distribution.

[0076] Step 303: When the pressure difference between adjacent pressure sub-areas is less than the pressure gradient threshold, the adjacent pressure sub-areas are divided into the same pressure area, and multiple target areas are divided according to the pressure center coordinates, the two-dimensional projection coordinates of the human skeleton, the position, area and pressure value distribution characteristics of the same pressure area.

[0077] In this embodiment, the pressure gradient threshold is a pre-set standard value. If the pressure difference between adjacent sub-regions is less than this threshold, it indicates that the pressure change between the two sub-regions is relatively gentle, without obvious abrupt changes. Therefore, they can be classified as the same pressure zone, which can better reflect the macroscopic characteristics of the pressure distribution. At the same time, for each identical pressure zone, its location on the mattress, area, and pressure distribution characteristics are recorded, such as the maximum pressure value, average pressure value, and uniformity of pressure distribution.

[0078] See also Figure 4 , Figure 4 A schematic diagram of dividing multiple target areas is provided in an embodiment of the present application. In some possible embodiments, the pressure center coordinates can be first superimposed and compared with the two-dimensional projection coordinates of the human skeleton. In the scenario where the human body lies flat on the mattress, the pressure center usually falls on the back or buttocks area of the torso. The two-dimensional projection coordinates of the human skeleton clearly show the position of the human skeleton on the mattress plane. By superimposing and comparing the pressure center coordinates with them, the specific position of the pressure center on the torso can be preliminarily locked based on the positional characteristics of the human skeleton.

[0079] Based on the determined pressure center, the complete trunk area is divided according to the position, area, and pressure value distribution characteristics of the same pressure area. On the mattress plane, some pressure areas that overlap with the projections of trunk bones such as ribs and sternum can be identified as trunk areas if their area and location (distance from the pressure center) conform to the distribution characteristics of the trunk on the mattress. In some embodiments, if the mean pressure value of the pressure area is at a high level and the pressure distribution pattern conforms to the characteristics of various parts of the trunk, such as a gradient change with relatively concentrated pressure distribution and higher pressure values on the back and relatively dispersed pressure and slightly lower pressure values on the waist, then it can be identified as a trunk area.

[0080] In some embodiments, the limb regions are distributed around the torso and correspond to the skeletal projections of the limbs. After determining the torso region, pressure regions that deviate from the pressure center and correspond to the direction of the skeletal projections of the limbs are searched. Based on the location, area, and pressure distribution characteristics of these pressure regions, regions that meet the characteristics of limbs are classified as limb regions. The head region is located away from the pressure center and corresponds to the location of the skull projection. Based on the set area threshold and pressure mean standard, the region where the head is located can be determined.

[0081] In the above embodiment, the overall pressure distribution data on the mattress is used to divide the human body into multiple target areas, such as the head area, torso area, and limb area. This can clearly determine the approximate position range of each part of the human body on the mattress, providing a basis for subsequently obtaining local pressure distribution data and accurately identifying the position of each part of the human body.

[0082] Step 203, obtain the local pressure distribution data of the local area of each target area through the air pressure sensor, and obtain the center point of each target area according to the local pressure distribution data. The local area of each target area is the area in the corresponding target area where the pressure data is greater than or equal to the threshold, and the center point of each target area is the position corresponding to the maximum value of the pressure data in the corresponding local area.

[0083] In an embodiment of the present application, the mattress includes multiple airbag units, and the air pressure sensor includes multiple air pressure sensor submodules, each of which is used to monitor the air pressure value of at least one airbag unit. The local pressure distribution data of the local area of each target area obtained by the air pressure sensor includes:

[0084] Increase the number of connections of the activated target air pressure sensor submodules in the local area of each target area, or dynamically connect the air pressure sensor submodule to the airbag unit of the local area through the air path switching device to adjust the monitoring position of the air pressure sensor submodule to obtain local pressure distribution data.

[0085] In some embodiments, the air pressure sensor submodules are the fundamental units of the entire air pressure sensor network. Each air pressure sensor submodule includes a pressure-sensitive element with independent pressure sensing and data acquisition capabilities. When subjected to pressure, it can monitor changes in air pressure within the airbag in real time and convert the pressure value into an electrical signal. The air pressure sensor submodules can be located within the airbag, distributed throughout the mattress, and can establish communication connections with the mattress or other submodules via an integrated communication module (such as WiFi, ZigBee, etc.).

[0086] In an embodiment of the present application, the mattress can use the overall pressure distribution data collected by the piezoelectric film sensor to determine areas within the target region where pressure data is greater than or equal to a threshold. For example, the pressure threshold can be set to 50 Pa. The target air pressure sensor submodule is an air pressure sensor submodule that is activated in a local area of the target region. The activated state means that the air pressure sensor submodule has transitioned from a standby or dormant state to an active state, enabling real-time monitoring and transmission of pressure data.

[0087] It will be appreciated that a greater number of target air pressure sensor submodules connected indicates a greater number of active air pressure sensor submodules capable of establishing communication with the mattress. In some embodiments, an air pressure sensor submodule can be dynamically connected to an airbag unit in a local area via an air path switching device to adjust the monitoring position of the air pressure sensor submodule. The following schematically illustrates two alternative implementations for obtaining local pressure distribution data.

[0088] See also Figure 5 , Figure 5 A schematic flow chart of obtaining local pressure distribution data provided in an embodiment of the present application is as follows: Figure 5 As shown, in some embodiments, the air pressure sensor is disposed inside the airbag unit, and increasing the number of connected target air pressure sensor submodules in an activated state in a local area of each target area may include the following steps:

[0089] Step 501 : Calculate the target number of air pressure sensor submodules that need to be added based on a preset sensor density threshold, the area of the local area, and the number of currently activated air pressure sensor submodules in the local area.

[0090] In an embodiment of the present application, the preset sensor density threshold represents the number of air pressure sensor submodules that should be activated per unit area. For example, a pressure sensor submodule can be disposed within each airbag unit, with each submodule corresponding to detecting the air pressure value of one airbag unit. The preset sensor density threshold can be set to a density threshold of 5 pressure sensor submodules per square decimeter. Based on the preset sensor density threshold and the area of the local area, the number of air pressure sensor submodules that should be activated in the local area is calculated. For example, the number of air pressure sensor submodules that should be activated in the local area can be obtained by multiplying the sensor density threshold by the area of the local area.

[0091] The target number of pressure sensor submodules to be added is calculated by calculating the difference between the desired number of activated pressure sensor submodules in the local area and the number of currently activated pressure sensor submodules in the local area. If the difference is zero or negative, the current sensor density has reached or exceeded the threshold and no additional pressure sensor submodules are needed.

[0092] Step 502 : activating a target number of air pressure sensor submodules in a dormant state in a local area, so that the number of connected target air pressure sensor submodules in the local area reaches a preset sensor density threshold.

[0093] In some embodiments, the dormant state refers to a low-power state in which the air pressure sensor submodule is inactive. In this dormant state, the submodule does not collect or transmit data, thereby conserving power and reducing unnecessary data processing. Optionally, after determining the target number of air pressure sensor submodules to be added, the mattress's microprocessor can send activation instructions via the communication module to the target number of air pressure sensor submodules in the dormant state within the local area. Upon receiving the activation instructions, the air pressure sensor submodules transition from the dormant state to the active state and begin collecting and transmitting pressure data.

[0094] In the above embodiment, by increasing the number of connections of the air pressure sensor sub-modules, the data acquisition density in the local area can be improved, thereby more accurately monitoring and analyzing the local pressure distribution data. In addition, activating the air pressure sensor sub-modules in the dormant state can more effectively utilize system resources. When high-density data acquisition is not required, some sub-modules can be put into a dormant state to save power.

[0095] In some embodiments, a plurality of air tubes are provided inside the mattress, the air tubes being connected to the airbag units. The air path switching device includes a plurality of micro air valves, which are used to control the passage of the air tubes. The air path switching device dynamically connects the air pressure sensor submodule to the airbag units in a local area to adjust the monitoring position of the air pressure sensor submodule, including:

[0096] determining a target airbag unit corresponding to a local area of each target area;

[0097] The micro air valve corresponding to the target airbag unit is controlled to be in an open state so that the air pressure sensor submodule is connected to the target airbag unit through the air pipe to adjust the monitoring position of the air pressure sensor submodule.

[0098] In an embodiment of the present application, the air pressure sensor submodule can also be located outside the airbag unit, for example, within a control box located at the foot of the bed. The air pressure sensor submodule is connected to the airbag unit via an air tube to monitor the air pressure inside the airbag unit. The air path switching device comprises a plurality of micro valves, each mounted on the airway of the air tube. By controlling the switching state of the micro valves, such as opening or closing, the air tube can be controlled to maintain continuity, thereby controlling the connection between the air pressure sensor submodule and the target airbag unit.

[0099] In some embodiments, each airbag unit within the mattress optionally has a fixed physical installation location, and the mattress can pre-store the spatial coordinate range corresponding to the location of each airbag unit. The coordinate range data for all airbag units can be stored in memory to form an "airbag-position mapping table." The location of a local area of each target region on the mattress can be converted into a specific coordinate range of that region in the mattress coordinate system. The mattress then uses the pre-stored "airbag-position mapping table" to compare the coordinate range of the local area with the coordinate ranges of each airbag unit. The airbag unit with the highest degree of coordinate overlap or that completely encompasses the local area is identified as the target airbag unit.

[0100] In some embodiments, each airbag unit may be connected to an independent air tube, and each air tube corresponds to a micro valve. This correspondence may be stored in a memory within the mattress. The mattress's microprocessor may send an electrical signal instruction to the micro valve corresponding to the target airbag unit to control the on / off state of the micro valve. For example, the electrical signal instruction may be transmitted via wires on a circuit board to a driver chip of the micro valve. The driver chip converts the control signal into an electric current that actuates the valve. The micro valve may be a solenoid valve. Upon receiving the driving current, the electromagnetic coil within the solenoid valve generates a magnetic field, attracting the valve core and opening the valve's inlet / outlet. At this point, the passage between the target airbag unit and the air tube is open, while the micro valves corresponding to non-target airbag units remain closed. This allows the air pressure sensor submodule to monitor the air pressure values of the airbag units corresponding to the local area of the target area to obtain local pressure distribution data.

[0101] In the above embodiment, by controlling the switching state of the micro air valve, the monitoring position of the air pressure sensor submodule can be dynamically adjusted, thereby achieving accurate monitoring of the pressure of a local area on the mattress. The monitoring position can also be adjusted in real time according to actual needs, which can improve the flexibility of local area pressure monitoring.

[0102] Step 204 : reconstructing various body parts of the human body according to the two-dimensional projection coordinates of the human skeleton and the center point of each target area, and identifying the positions of various body parts of the human body on the mattress.

[0103] In the embodiment of the present application, the center point of each target area is the location of the maximum pressure data within the local area of each target area. It is understandable that when a human body contacts the mattress surface, key skeletal points of the human body, such as bony protrusions such as shoulders, elbows, hips, and knees, will generate greater pressure values than surrounding areas due to the lack of soft tissue cushioning. Therefore, in some embodiments, the method of reconstructing various body parts of the human body based on the two-dimensional projection coordinates of the human skeleton and the center point of each target area and identifying the position of each body part of the human body on the mattress includes:

[0104] Map the position of the center point of each target area on the mattress to the position of the key points of the human skeleton;

[0105] According to the two-dimensional projection coordinates of the human skeleton and the positions of the key points of the skeleton, the various body parts of the human body are reconstructed to obtain the positions of the various body parts of the human body on the mattress.

[0106] In some embodiments, for example, the position of the center point of the torso area on the mattress may include the positions of the acromion, thoracic vertebrae, lumbar vertebrae and other skeletal key points on the mattress; the position of the center point of the limb area on the mattress may include the positions of the elbow joint, wrist joint, hip joint, knee joint, ankle joint and other skeletal key points on the mattress; the main skeletal key point of the head area is the occipital protuberance.

[0107] It can be understood that by mapping the position of the center point of each target area on the mattress to the position of the skeletal key points of the human skeleton, the pressure data on the mattress can be matched with the skeletal key points of the human skeleton, thereby determining the specific positions of the key points of various parts of the human body on the mattress.

[0108] In some embodiments, the relative positions and postures of various body parts can be determined using the known positions of skeletal key points and the two-dimensional projection coordinates of the human skeleton, thereby enabling complete reconstruction of the body's posture on the mattress. Based on the specific positions of the skeletal key points on the mattress and the body's posture, the specific positions of various body parts on the mattress can be further determined. For example, the position of the calf on the mattress can be determined based on the position of the knee and ankle joints on the mattress and the body's posture.

[0109] In the above embodiment, by matching the pressure data with the key points of the human skeleton, the accuracy of the position recognition of each part of the human body on the mattress can be improved, thereby more accurately identifying the position and posture of each part of the human body on the mattress.

[0110] In some other embodiments, the human body part recognition method further includes:

[0111] The three-dimensional contour data, overall pressure distribution data, and local pressure distribution data are input into a multimodal data fusion model to obtain the posture of the human body on the mattress and the positions of various parts of the human body on the mattress surface. The multimodal data fusion model is obtained by training an initial model based on the sample three-dimensional contour data, the sample overall pressure distribution data, the sample local pressure distribution data, and the corresponding human posture annotation information and the position annotation information of various parts of the human body.

[0112] In some embodiments, the initial model for multimodal data fusion can be a deep learning model, such as a convolutional neural network, a recurrent neural network, or a Transformer architecture. Sample three-dimensional contour data, sample overall pressure distribution data, and sample local pressure distribution data are input into the initial model. These data can be input into the model separately through different input channels or fused at a certain stage of the model. The initial model is trained using corresponding human posture annotation information and position annotation information of various human body parts as supervisory signals, and the model parameters are optimized by minimizing the difference between the predicted values and the annotated values.

[0113] Through training with a large amount of sample data, the model learns the mapping relationship from multimodal data to human posture and the positions of various body parts. For example, the multimodal data fusion model outputs predicted results of the human body posture on the mattress based on the input multimodal data. These results can include the overall posture of the human body, such as supine, side-lying, prone, etc. The multimodal data fusion model can also simultaneously output predicted results of the positions of various human body parts on the mattress surface. These results can include the specific positions of various parts such as the head, shoulders, waist, back, buttocks, thighs, calves, and arms.

[0114] In the above embodiment, by combining three-dimensional contour data, overall pressure distribution data, and local pressure distribution data, the model can obtain information about the human body on the mattress from multiple angles, thereby more comprehensively and accurately identifying the posture of the human body on the mattress and the position of various body parts on the mattress.

[0115] In some embodiments, the mattress further comprises a temperature sensor array, and the method further comprises:

[0116] Receive mattress usage scenario information input by the user. If the scenario information is a medical care scenario, obtain temperature data of the contact area between the human body part and the mattress through the temperature sensor array, and evaluate the blood circulation status of the human body part based on the temperature data and pressure data;

[0117] When an abnormality in the blood circulation state is detected, the human body part with the abnormal blood circulation state is determined based on the recognition results of the positions of various human body parts on the mattress surface, and a prompt message is output.

[0118] In some embodiments, users can input mattress usage scenario information through an electronic device or application associated with the mattress. For example, users can select "medical care scenario," "sleep scenario," "rehabilitation training scenario," etc. In the medical care scenario, the temperature sensor array installed on the mattress can collect real-time temperature data from the area of contact between the human body and the mattress, and transmit the data to the mattress's microprocessor for processing and analysis.

[0119] In some embodiments, the step of evaluating the blood circulation status of a human body part based on the temperature data and the pressure data includes:

[0120] Obtain the temperature change rate of the temperature data within a preset time period, as well as the pressure value and pressure duration corresponding to the pressure data;

[0121] When the temperature change rate is greater than or equal to the first threshold, or the pressure value is greater than or equal to the second threshold and the pressure duration is greater than or equal to the third threshold, it is determined that the blood circulation state of the human body part is abnormal.

[0122] It's understandable that when blood circulation is obstructed, the corresponding body part may experience a drop in temperature. Therefore, by monitoring the rate of temperature change, blood circulation abnormalities can be promptly detected. Furthermore, if the pressure in the area of contact between the body and the mattress is greater than or equal to the second threshold and the duration of the pressure is greater than or equal to the third threshold, it indicates that the blood vessels in the corresponding body part may have been excessively compressed for a prolonged period, potentially leading to blood circulation abnormalities.

[0123] By adopting the above embodiment, it is possible to timely detect and identify parts of the human body with abnormal blood circulation, and output prompt information, so that medical staff can take timely measures, such as adjusting the patient's posture or performing local massage, which can effectively improve the quality of care.

[0124] See also Figure 6 , Figure 6 A schematic diagram of the structure of the mattress provided in the embodiment of the present application is shown in FIG. Figure 6 As shown, the mattress 600 may include a first processing module 601, a second processing module 602, a third processing module 603 and an identification module 604, wherein:

[0125] The first processing module 601 is configured to obtain three-dimensional contour data of a human body on the mattress using a laser radar sensor, and extract two-dimensional projection coordinates of human bones based on the three-dimensional contour data;

[0126] A second processing module 602 is configured to obtain overall pressure distribution data of the mattress using a piezoelectric film sensor, and divide the mattress into a plurality of target areas based on the overall pressure distribution data and the two-dimensional projection coordinates of the human skeleton, wherein the plurality of target areas include a head area, a torso area, and a limb area;

[0127] A third processing module 603 is configured to obtain local pressure distribution data of a local area of each target area through a pressure sensor, and obtain a center point of each target area based on the local pressure distribution data, wherein the local area of each target area is an area in the corresponding target area where the pressure data is greater than or equal to a threshold, and the center point of each target area is a position corresponding to the maximum value of the pressure data in the corresponding local area;

[0128] The identification module 604 is used to reconstruct the various body parts of the human body according to the two-dimensional projection coordinates of the human skeleton and the center point of each target area, and identify the position of each body part of the human body on the mattress.

[0129] In some embodiments, the air pressure sensor includes multiple air pressure sensor submodules, and the third processing module 603 is specifically configured to:

[0130] Increase the number of connections of the activated target air pressure sensor submodules in the local area of each target area, or dynamically connect the air pressure sensor submodule to the airbag unit of the local area through an air path switching device to adjust the monitoring position of the air pressure sensor submodule to obtain the local pressure distribution data.

[0131] In some embodiments, the air pressure sensor is disposed inside the airbag unit, and the third processing module 603 is specifically configured to:

[0132] Calculating a target number of air pressure sensor submodules to be added based on a preset sensor density threshold, the area of the local area, and the number of currently activated air pressure sensor submodules in the local area;

[0133] A target number of air pressure sensor submodules in a dormant state in the local area are activated so that the number of connection of the target air pressure sensor submodules in the local area reaches the preset sensor density threshold.

[0134] In some embodiments, the mattress includes a plurality of air tubes, the air tubes being in communication with the airbag units, the air path switching device includes a plurality of micro air valves, the micro air valves being used to control the passage of the air tubes, and the third processing module 603 is specifically configured to:

[0135] determining a target airbag unit corresponding to a local area of each target area;

[0136] The micro air valve corresponding to the target airbag unit is controlled to be in an open state, so that the air pressure sensor submodule is connected to the target airbag unit through the air pipe, so as to adjust the monitoring position of the air pressure sensor submodule.

[0137] In some embodiments, the second processing module 602 is specifically configured to:

[0138] Gridding the overall pressure distribution data to divide it into multiple pressure sub-regions, and calculating the pressure value and area of each pressure sub-region;

[0139] Calculating the pressure center coordinates of the overall pressure distribution data using a weighted average algorithm based on the pressure value and area of each pressure sub-region;

[0140] When the pressure difference between adjacent pressure sub-areas is less than the pressure gradient threshold, the adjacent pressure sub-areas are divided into the same pressure area, and the multiple target areas are divided according to the pressure center coordinates, the two-dimensional projection coordinates of the human skeleton, the position, area and pressure value distribution characteristics of the same pressure area.

[0141] In some embodiments, the identification module 604 is specifically configured to:

[0142] Mapping the position of the center point of each target area on the mattress to the position of the key points of the human skeleton;

[0143] According to the two-dimensional projection coordinates of the human skeleton and the positions of the key points of the skeleton, various body parts of the human body are reconstructed to obtain the positions of the various body parts of the human body on the mattress.

[0144] In some embodiments, the first processing module 601 is specifically configured to:

[0145] Mapping each data point in the three-dimensional contour data to the mattress plane coordinate system through projection transformation to obtain corresponding two-dimensional projection data;

[0146] The two-dimensional projection data is input into a pre-trained human posture estimation model, the human skeleton is extracted, and the two-dimensional projection coordinates of the human skeleton are obtained. The human posture estimation model is obtained by training an initial model based on the human posture and the human skeleton corresponding to the human posture.

[0147] The description of the above device embodiment is similar to the description of the above method embodiment and has similar beneficial effects as the method embodiment. For technical details not disclosed in the device embodiment of this application, please refer to the description of the method embodiment of this application for understanding.

[0148] It should be noted that in the embodiments of this application Figure 6 The module division of mattress 600 shown is schematic and represents only one logical functional division; actual implementations may employ different division methods. Furthermore, the functional units in various embodiments of this application may be integrated into a single processing unit, physically exist separately, or two or more units may be integrated into a single unit. These integrated units may be implemented in hardware or software functional units, or may be implemented using a combination of software and hardware.

[0149] It should be noted that, in the embodiment of the present application, if the above method is implemented in the form of a software function module and sold or used as an independent product, it can also 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 relevant technology 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 for enabling an electronic device to execute all or part of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk. In this way, the embodiment of the present application is not limited to any specific combination of hardware and software.

[0150] See also Figure 7 , Figure 7 This is a structural block diagram of the electronic device provided in the embodiment of the present application. Figure 7 As shown, the electronic device 700 may include: a processor 710 , a memory 720 , and a bus 730 .

[0151] The processor 710 calls the executable program code stored in the memory 720 to execute any of the human body part recognition methods disclosed in the embodiments of the present application. Figure 7 The electronic device structure shown in the figure does not constitute a limitation to the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0152] In the embodiments of the present application, the processor 710 may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and may implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present application. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present application may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0153] The memory 720 can be used to store software programs and modules. The processor 710 executes the various functional applications and data processing of the electronic device by running the software programs and modules stored in the memory 720. The memory 720 may mainly include a program storage area and a data storage area. The program storage area may store an operating system, at least one application required for a function, and the like; the data storage area may store data generated based on the use of the electronic device. In addition, the memory 720 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state memory device.

[0154] In the embodiment of the present application, the processor 710 and the memory 720 are connected via a bus 730. Figure 7 The connections between the other components are shown in bold lines, which are only for illustration and not intended to be limiting. The bus can be divided into address bus, data bus, control bus, etc. Figure 7 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0155] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method provided in the above embodiment are implemented.

[0156] An embodiment of the present application provides a computer program product containing instructions, which, when executed on a computer, enables the computer to execute the steps of the method provided in the above method embodiment.

[0157] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a ROM, or the like.

[0158] It should be noted that the description of the above storage medium and device embodiments is similar to the description of the above method embodiments and has similar beneficial effects as the method embodiments. For technical details not disclosed in the storage medium, storage medium, and device embodiments of this application, please refer to the description of the method embodiments of this application for understanding.

[0159] It should be understood that "one embodiment" or "an embodiment" or "some embodiments" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" or "in some embodiments" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above-mentioned serial numbers of the embodiments of the present application are for description only and do not represent the advantages and disadvantages of the embodiments. The above description of the various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other. For the sake of brevity, they will not be repeated here.

[0160] The term "and / or" in this article is only a description of the association relationship between associated objects, indicating that there can be three relationships. For example, object A and / or object B can mean: object A exists alone, object A and object B exist at the same time, and object B exists alone.

[0161] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0162] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or modules can be electrical, mechanical or other forms.

[0163] The modules described above as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules; they may be located in one place or distributed across multiple network units; some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0164] In addition, all functional modules in the embodiments of the present application can be integrated into one processing unit, or each module can be a separate unit, or two or more modules can be integrated into one unit; the above-mentioned integrated modules can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0165] Those skilled in the art will understand that all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM), magnetic disks or optical disks, and other media that can store program codes.

[0166] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application, or the part that contributes to the relevant technology, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling an electronic device to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks or optical disks.

[0167] The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0168] The features disclosed in the several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0169] The features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0170] The above is merely an embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for identifying human body parts, characterized in that: Applied to a mattress, the mattress includes a piezoelectric film sensor and an air pressure sensor, a lidar sensor is disposed above the mattress, the lidar sensor is communicatively connected to the mattress, and the method includes: Acquiring three-dimensional contour data of a human body on the mattress by the laser radar sensor, and extracting two-dimensional projection coordinates of the human skeleton based on the three-dimensional contour data; Acquiring overall pressure distribution data of the mattress through the piezoelectric film sensor, and dividing the mattress into multiple target areas according to the overall pressure distribution data and the two-dimensional projection coordinates of the human skeleton, wherein the multiple target areas include a head area, a torso area, and a limb area; Acquire local pressure distribution data of a local area of each target area through the air pressure sensor, and obtain the center point of each target area according to the local pressure distribution data, wherein the local area of each target area is an area in the corresponding target area where the pressure data is greater than or equal to a threshold value, and the center point of each target area is a position corresponding to the maximum value of the pressure data in the corresponding local area; According to the two-dimensional projection coordinates of the human skeleton and the center point of each target area, the various body parts of the human body are reconstructed, and the positions of the various body parts of the human body on the mattress are identified.

2. The method according to claim 1, characterized in that The mattress includes a plurality of airbag units, the air pressure sensor includes a plurality of air pressure sensor submodules, each air pressure sensor submodule is used to monitor the air pressure value of at least one airbag unit, and the local pressure distribution data of a local area of each target area is obtained by the air pressure sensor, including: Increase the number of connections of the activated target air pressure sensor submodules in the local area of each target area, or dynamically connect the air pressure sensor submodule to the airbag unit of the local area through an air path switching device to adjust the monitoring position of the air pressure sensor submodule to obtain the local pressure distribution data.

3. The method according to claim 2, characterized in that The air pressure sensor is disposed inside the airbag unit, and increasing the number of connected target air pressure sensor submodules in an activated state in a local area of each target area includes: Calculating a target number of air pressure sensor submodules to be added based on a preset sensor density threshold, the area of the local area, and the number of currently activated air pressure sensor submodules in the local area; A target number of air pressure sensor submodules in a dormant state in the local area are activated so that the number of connection of the target air pressure sensor submodules in the local area reaches the preset sensor density threshold.

4. The method according to claim 2, characterized in that The mattress includes a plurality of air tubes connected to the airbag units. The air path switching device includes a plurality of micro air valves for controlling the passage of the air tubes. The air pressure sensor submodule is dynamically connected to the airbag units in the local area through the air path switching device to adjust the monitoring position of the air pressure sensor submodule, including: determining a target airbag unit corresponding to a local area of each target area; The micro air valve corresponding to the target airbag unit is controlled to be in an open state, so that the air pressure sensor submodule is connected to the target airbag unit through the air pipe, so as to adjust the monitoring position of the air pressure sensor submodule.

5. The method according to claim 1, wherein The step of dividing a plurality of target areas according to the overall pressure distribution data and the two-dimensional projection coordinates of the human skeleton includes: Gridding the overall pressure distribution data to divide it into multiple pressure sub-regions, and calculating the pressure value and area of each pressure sub-region; Calculating the pressure center coordinates of the overall pressure distribution data using a weighted average algorithm based on the pressure value and area of each pressure sub-region; When the pressure difference between adjacent pressure sub-areas is less than the pressure gradient threshold, the adjacent pressure sub-areas are divided into the same pressure area, and the multiple target areas are divided according to the pressure center coordinates, the two-dimensional projection coordinates of the human skeleton, the position, area and pressure value distribution characteristics of the same pressure area.

6. The method according to claim 1, wherein The reconstructing each body part of the human body according to the two-dimensional projection coordinates of the human skeleton and the center point of each target area, and identifying the position of each body part of the human body on the mattress, includes: Mapping the position of the center point of each target area on the mattress to the position of the key points of the human skeleton; According to the two-dimensional projection coordinates of the human skeleton and the positions of the key points of the skeleton, various body parts of the human body are reconstructed to obtain the positions of the various body parts of the human body on the mattress.

7. The method according to claim 1, characterized in that The step of extracting the two-dimensional projection coordinates of the human skeleton according to the three-dimensional contour data comprises: Mapping each data point in the three-dimensional contour data to the mattress plane coordinate system through projection transformation to obtain corresponding two-dimensional projection data; The two-dimensional projection data is input into a pre-trained human posture estimation model, the human skeleton is extracted, and the two-dimensional projection coordinates of the human skeleton are obtained. The human posture estimation model is obtained by training an initial model based on the human posture and the human skeleton corresponding to the human posture.

8. A mattress, characterized in that: The mattress includes a piezoelectric film sensor and an air pressure sensor. A laser radar sensor is disposed above the mattress and is in communication with the mattress. The mattress includes: a first processing module, configured to obtain three-dimensional contour data of a human body on the mattress through the laser radar sensor, and extract two-dimensional projection coordinates of human bones based on the three-dimensional contour data; a second processing module, configured to obtain overall pressure distribution data of the mattress through the piezoelectric film sensor, and divide the mattress into a plurality of target areas according to the overall pressure distribution data and the two-dimensional projection coordinates of the human skeleton, the plurality of target areas including a head area, a torso area, and a limb area; a third processing module, configured to obtain local pressure distribution data of a local area of each target area through the air pressure sensor, and obtain a center point of each target area based on the local pressure distribution data, wherein the local area of each target area is an area in the corresponding target area where the pressure data is greater than or equal to a threshold, and the center point of each target area is a position corresponding to the maximum value of the pressure data in the corresponding local area; The recognition module is used to reconstruct various body parts of the human body according to the two-dimensional projection coordinates of the human skeleton and the center point of each target area, and to identify the position of each body part of the human body on the mattress.

9. A mattress, characterized in that: include: a memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images