Control and cleaning methods, devices, and electronic equipment for nursing beds

Abstract

The embodiments of the present disclosure disclose a control and cleaning method, device, and electronic device for use in a nursing bed. A specific implementation of the method includes: determining the body posture information corresponding to the target object; collecting real-time temperature signals and real-time humidity signals through the temperature and humidity sensors included in the nursing bed; determining temperature and humidity control information based on the real-time temperature signals and real-time humidity signals; controlling the temperature and humidity adjustment device included in the nursing bed to adjust the temperature and humidity based on the temperature and humidity control information; in response to obtaining an active control signal, performing signal analysis on the active control signal; controlling the nursing bed; in response to reaching a predetermined time point, initiating a body cleaning reminder for the target object to a remote monitoring terminal, wherein the nursing bed also includes an auxiliary cleaning device for the patient. This implementation effectively solves the problem that medical staff find it difficult to effectively and timely monitor the patient's posture due to the unbalanced doctor-patient ratio, and the problem of body temperature imbalance caused by patients being bedridden for a long time.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the medical field and the computer field, and more particularly to a control and cleaning method, device, and electronic device for a nursing bed. Background Art

[0002] Nursing beds are specialized beds designed for patients requiring long-term bed rest or medical care. However, the imbalance in the doctor-patient ratio makes it difficult for medical staff to effectively and promptly monitor patients' posture. Furthermore, prolonged bed rest can easily lead to temperature imbalances and other issues.

[0003] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0004] The content of this disclosure is used to briefly introduce concepts that will be described in detail in the detailed description section below. The content of this disclosure is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0005] Some embodiments of the present disclosure propose control and cleaning methods, devices, and electronic equipment for nursing beds to solve the technical problems mentioned in the above background technology section.

[0006] In a first aspect, some embodiments of the present disclosure provide a control and cleaning method for a nursing bed, the method comprising: in response to a target object being located in an area to be detected, collecting a real-time image group sequence, wherein the area to be detected is the area where the nursing bed is located, the target object wears a wireless radio frequency bracelet bound to the identity of the nursing bed, and the real-time image group comprises: a real-time infrared image and a real-time non-infrared image; according to the real-time image group sequence, determining body posture information corresponding to the target object, wherein the body posture information comprises: a torso key point information set and a posture type; in response to the body posture information indicating that the posture of the target object is normal and the posture type is a lying posture, the temperature control unit included in the nursing bed is used to control the posture of the target object. A humidity sensor collects real-time temperature signals and real-time humidity signals; determines temperature and humidity control information based on the above-mentioned real-time temperature signals and the above-mentioned real-time humidity signals, wherein the above-mentioned temperature and humidity control information includes: control mode, control area and control gear; controls the temperature and humidity adjustment device included in the above-mentioned nursing bed to adjust the temperature and humidity according to the above-mentioned temperature and humidity control information; in response to obtaining an active control signal, performs signal analysis on the above-mentioned active control signal to generate nursing bed control information; controls the above-mentioned nursing bed according to the above-mentioned nursing bed control information; in response to reaching a predetermined time point, initiates a body cleaning reminder for the above-mentioned target object to the remote monitoring terminal, wherein the nursing bed also includes an auxiliary cleaning device for the patient.

[0007] In a second aspect, some embodiments of the present disclosure provide a control and cleaning device for a nursing bed, the device comprising: a first acquisition unit, configured to acquire a real-time image group sequence in response to a target object being located in an area to be detected, wherein the area to be detected is the area where the nursing bed is located, the target object wears a wireless radio frequency bracelet bound to the identity of the nursing bed, and the real-time image group comprises: a real-time infrared image and a real-time non-infrared image; a first determination unit, configured to determine the body posture information corresponding to the target object according to the real-time image group sequence, wherein the body posture information comprises: a torso key point information set and a posture type; a second acquisition unit, configured to acquire real-time temperature and humidity through the temperature and humidity sensor included in the nursing bed in response to the body posture information indicating that the posture of the target object is normal and the posture type is a lying posture. signal and real-time humidity signal; a second determination unit is configured to determine temperature and humidity control information according to the above-mentioned real-time temperature signal and the above-mentioned real-time humidity signal, wherein the above-mentioned temperature and humidity control information includes: control mode, control area and control gear; a first control unit is configured to control the temperature and humidity adjustment device included in the above-mentioned nursing bed to adjust the temperature and humidity according to the above-mentioned temperature and humidity control information; a signal analysis unit is configured to perform signal analysis on the above-mentioned active control signal in response to obtaining the active control signal to generate nursing bed control information; a second control unit is configured to control the above-mentioned nursing bed according to the above-mentioned nursing bed control information; a body cleaning reminder unit is configured to initiate a body cleaning reminder for the above-mentioned target object to the remote monitoring terminal in response to reaching a predetermined time point, wherein the nursing bed also includes an auxiliary cleaning device for the patient.

[0008] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by one or more processors, the one or more processors implement the method described in any implementation of the first aspect above.

[0009] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation of the first aspect is implemented.

[0010] The above-described embodiments of the present disclosure have the following beneficial effects: Through the control and cleaning methods applied to nursing beds according to some embodiments of the present disclosure, effective and timely monitoring of patient posture is achieved, while effectively addressing the problem of body temperature imbalance caused by long-term bed rest. Specifically, the present disclosure first captures a real-time image sequence in response to a target subject being located within a detection area. The detection area is the area surrounding the nursing bed, and the target subject is wearing a wireless radio frequency bracelet associated with the bed. The real-time image sequence includes real-time infrared images and real-time non-infrared images. In practice, the real-time image sequence is captured only when the target subject (patient) is within the detection area. This avoids capturing images when the target subject is not within the detection area, reducing the number of images captured and the subsequent image processing required. Secondly, based on the real-time image sequence, body posture information corresponding to the target subject is determined. The body posture information includes a set of torso key point information and a posture type. In practice, since the target subject may be covered by objects (e.g., bedding) when lying on the nursing bed, body posture recognition is performed by combining infrared and non-infrared images. Furthermore, in response to the body posture information indicating that the target subject is in a normal posture and the posture type is a reclining posture, a real-time temperature signal and a real-time humidity signal are collected via a temperature and humidity sensor included in the nursing bed. Furthermore, temperature and humidity control information is determined based on the real-time temperature and humidity signals, wherein the temperature and humidity control information includes a control mode, a control area, and a control gear. Then, based on the temperature and humidity control information, a temperature and humidity adjustment device included in the nursing bed is controlled to adjust the temperature and humidity. By combining the temperature and humidity sensor, whether the target subject experiences body temperature imbalance while in a reclining position on the nursing bed is analyzed, and the temperature and humidity are adjusted accordingly. Furthermore, in response to obtaining an active control signal, the active control signal is analyzed to generate nursing bed control information. Then, the nursing bed is controlled based on the nursing bed control information. Finally, in response to reaching a predetermined time point, a body cleaning reminder for the target subject is sent to a remote monitoring terminal, wherein the nursing bed also includes an auxiliary cleaning device for the patient. This approach effectively addresses the issue of medical staff being unable to monitor patients’ posture effectively and promptly due to the imbalance in the doctor-patient ratio. It also effectively addresses issues such as body temperature imbalances caused by prolonged bed rest. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0012] Figure 1 is a flow chart of some embodiments of a control and cleaning method applied to a nursing bed according to the present disclosure;

[0013] Figure 2 This is a schematic diagram of the position relationship between the camera and the nursing bed;

[0014] Figure 3 It is a schematic diagram of body postures of different posture types;

[0015] Figure 4 It is a schematic diagram of the local model structure of the dual-path image feature extraction model;

[0016] Figure 5 It is a schematic diagram of the process of determining the basic detection area;

[0017] Figure 6 Schematic diagram of the model structure of the body posture information prediction model;

[0018] Figure 7 This is a schematic diagram of the control effect of the nursing bed in heating mode and ventilation mode;

[0019] Figure 8 is a schematic structural diagram of some embodiments of a control and cleaning device applied to a nursing bed according to the present disclosure;

[0020] Figure 9 It is a structural diagram of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0021] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0022] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other.

[0023] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0024] It should be noted that the modifications of "one" and "multiple" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, they should be understood as "one or more".

[0025] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0026] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0027] refer to Figure 1 , shows a process 100 of some embodiments of the control and cleaning method for a nursing bed according to the present disclosure. The control and cleaning method for a nursing bed includes the following steps:

[0028] Step 101: In response to a target object being located in a region to be detected, a real-time image group sequence is acquired.

[0029] In some embodiments, an executing entity (e.g., a computing device) of a control and cleaning method for a nursing bed can capture a sequence of real-time image groups in response to a target object being located within a detection area. In practice, the executing entity can control a camera to capture the sequence of real-time image groups. The detection area is the area where the nursing bed is located. In practice, the camera can be a wide-angle camera with a fixed shooting angle. Given a fixed shooting angle, the relative position of the nursing bed and the camera is fixed, so the detection area can be pre-set within the camera's image. The target object wears a wireless radio frequency bracelet that is associated with the nursing bed. The target object can be a patient using the nursing bed. The wireless radio frequency bracelet can be worn on the target object's wrist. Specifically, the wireless radio frequency bracelet can include two radio frequency transmitters. The real-time images in the sequence of real-time image groups are images of the detection area. The real-time image groups include real-time infrared images and real-time non-infrared images. In practice, the camera can include a conventional wide-angle camera and an infrared camera to capture both non-infrared and infrared images. Furthermore, the camera can include a radio frequency receiver for receiving radio frequency signals transmitted by the radio frequency transmitter included in the wireless radio frequency bracelet.

[0030] It should be noted that, in practice, since there are a large number of nursing beds inside the hospital, there is a need for monitoring of multiple nursing beds. In order to cope with the corresponding large computing power requirements, the above-mentioned computing device can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or it can be implemented as a single server or a single terminal device. When the computing device is embodied as software, it can be installed in the hardware devices listed above. It can be implemented as multiple software or software modules for providing distributed services, for example, or it can be implemented as a single software or software module. No specific limitation is made here.

[0031] As an example, see the schematic diagram of the position relationship between the camera and the nursing bed shown in Figure 2, where: Figure 2 The illustrated nursing ward may include: nursing bed A, nursing bed B, nursing bed C, and nursing bed D. The camera can be a wide-angle camera so that the camera's image capture includes nursing bed A, nursing bed B, nursing bed C, and nursing bed D, effectively reducing camera deployment costs. Specifically, since the positional relationship between the nursing bed and the camera is relatively fixed, a detection area is pre-set for each nursing bed within the camera's image capture. Furthermore, given the potential for multiple beds within a nursing ward, it's difficult to effectively maintain a fixed relationship between nursing beds and patients. Conventional image recognition methods require the capture of numerous patient facial images, resulting in extremely high maintenance costs. Given that patients are required to wear identification bracelets upon admission, a wireless RF bracelet, combined with the camera's included RF receiver, can be used to associate the patient's identity with the nursing bed and detect whether the target object is within the detection area. Specifically, since the wireless RF bracelet contains two RF transmitters, the target object can be located based on the time difference between the RF signals received by the RF receivers (at least three) from the two transmitters. On this basis, since the camera position is fixed and the camera is pre-calibrated, the three-dimensional coordinates of the identified target object in the geodetic coordinate system can be mapped to the image coordinate system through conversion between the geodetic coordinate system and the image coordinate system, so as to determine whether the target object is located in the area to be detected.

[0032] Step 102: Determine body posture information corresponding to the target object based on the real-time image group sequence.

[0033] In some embodiments, the execution subject can determine the body posture information corresponding to the target object based on the real-time image group sequence. The body posture information includes: a torso key point information set and a posture type. The posture type represents the body posture of the target object. The trunk key point information in the trunk key point information set includes: key point coordinates and torso type. The torso type represents the type of the trunk corresponding to the trunk key point of the trunk key point information. In practice, the posture types include: supine posture type, left side lying posture type, right side lying posture type, left curled side lying posture type, and right curled side lying posture type. In practice, the YOLOv8 model can be used to determine the body posture information corresponding to the target object based on the real-time image group sequence.

[0034] For example, see Figure 3 The diagram shows body postures of different posture types, wherein body postures of supine posture type, left side lying posture type, right side lying posture type, left curled up side lying posture type, and right curled up side lying posture type are shown.

[0035] In some optional implementations of some embodiments, the execution subject determines the body posture information corresponding to the target object based on the real-time image group sequence, including:

[0036] In the first step, for each real-time image group in the real-time image group sequence, the following first processing step is performed:

[0037] In the first sub-step, image features are extracted from the real-time infrared image and the real-time non-infrared image included in the real-time image group using a dual-path image feature extraction model to generate an image feature map group.

[0038] The image feature map group includes an infrared image feature map and a non-infrared image feature map. The infrared image feature map and the non-infrared image feature map have the same feature map size. Specifically, the dual-path image feature extraction model includes an image feature extraction model A and an image feature extraction model B. The model structures of image feature extraction model A and image feature extraction model B are consistent and symmetrical. The input of image feature extraction model A is a real-time infrared image, and the output is an infrared image feature map. The input of image feature extraction model B is a real-time non-infrared image, and the output is a non-infrared image feature map.

[0039] For example, see Figure 4The schematic diagram of the local model structure of the dual-path image feature extraction model shown in the figure shows the model structure of the image feature extraction model A. Specifically, the image feature extraction model A includes: image feature extraction model A1 and image feature extraction model A2. The image feature extraction model A1 adopts a conventional convolutional neural network structure. Specifically, the image feature extraction model A1 includes: convolution layer A1, convolution layer A2, convolution layer A3, convolution layer A4 and convolution layer A5. The input of convolution layer A1 is a real-time infrared image. The connection relationship between the convolution layer and the fully connected layer included in the image feature extraction model A1 is as follows: Figure 4 As shown. Convolutional layer A2, convolutional layer A3, convolutional layer A4 and convolutional layer A5 are all residual convolutional layers, and all use ReLU activation function. Image feature extraction model A2 adopts Unet neural network structure. Specifically, image feature extraction model A2 adopts conventional convolutional neural network structure including: convolutional layer A6, convolutional layer A7, convolutional layer A8, convolutional layer A9, convolutional layer A10, convolutional layer A11, convolutional layer A12, convolutional layer A13, convolutional layer A14, convolutional layer A15, convolutional layer A16 and fully connected layer A17. The connection relationship between the convolutional layer and the fully connected layer included in the image feature extraction model A2 is shown as follows: Figure 4 As shown. Fully connected layer A17 is used to superimpose the features output by convolutional layers A5 and A16. Furthermore, the model structure of image feature extraction model B is consistent with that of image feature extraction model A. Image feature extraction model B comprises: image feature extraction model B1 and image feature extraction model B2. Specifically, the dual-path image feature extraction model described above, as an inventive feature of the present disclosure, shares parameters between the convolutional layers corresponding to image feature extraction models A1 and B1. Parameters are shared between the convolutional layers corresponding to image feature extraction models A2 and B2. Taking image feature extraction model A as an example, a lightweight image feature extraction model A1 is used to capture fine image features in the image, while a lightweight image feature extraction model A2 is used to capture spatial contextual information in the image. Furthermore, considering that when the target object is covered (e.g., covered by bedding), recognition results are poor when only real-time non-infrared images are used for recognition. Therefore, real-time infrared images are recognized in conjunction with image feature extraction model B, and parameters are shared between the corresponding models, thereby reducing the number of model parameters.

[0040] The second sub-step is to construct a basic detection area based on the area to be detected.

[0041] The corner points of the area to be detected are located on the area boundary of the basic detection area.

[0042] In practice, see further Figure 2The position relationship diagram of the camera and the nursing bed is shown in the figure. Since the camera needs to shoot at a certain tilt angle to collect images containing the nursing bed, the nursing bed in the detection area exists at a certain tilt angle in the real-time infrared image and the real-time non-infrared image. For details, please refer to Figure 5 The following figure shows a schematic diagram of the process for determining the basic detection area. In this diagram, the nursing bed within the detection area 501 within the real-time non-infrared image 502 is tilted at a certain angle. Based on this, the four image boundaries of the real-time non-infrared image 502 are horizontally shifted inward until they contact the corners of the detection area 501. At this point, the area enclosed by the image boundaries is defined as the basic detection area 503.

[0043] The third sub-step is to construct a set of candidate detection areas based on the above basic detection areas and the preset window magnification ratio.

[0044] Among them, the above-mentioned basic detection area is a regional subset of the candidate detection area in the above-mentioned candidate detection area set. In practice, the window magnification ratio refers to the magnification ratio of the basic detection area along the diagonal. For example, the window magnification ratio can be 1%. In practice, for target detection, it is usually necessary to construct a large number of detection frames and perform target detection on each detection frame. However, the present disclosure takes into account the need to perform body posture detection of the target object in combination with the area to be detected, and the target object is already located in the area to be detected. Therefore, a basic detection area constructed based on the area to be detected can be used to construct a small number of candidate detection area sets, thereby reducing the number of detection frames and improving the detection speed.

[0045] The fourth sub-step is to determine, for each candidate detection area in the candidate detection area set, the area confidence of the candidate detection area based on the candidate detection area, the infrared image feature map and the non-infrared image feature map included in the image feature map group.

[0046] Among them, the above-mentioned regional confidence represents the confidence of whether the candidate detection area contains the target object. Specifically, the above-mentioned execution entity can determine the regional confidence of the candidate detection area based on the local features of the candidate detection area in the infrared image feature map and the non-infrared image feature map through a binary classifier. Specifically, the binary classifier is used to determine whether the candidate detection area contains the target object. During the model training stage, the binary classifier and the two-way image feature extraction model can be trained together. Specifically, the binary classifier and the two-way image feature extraction model are trained as a whole using training samples marked with the location of the target object in a supervised training method.

[0047] The fifth sub-step is to set the feature values ​​outside the target detection area in the infrared image feature map included in the above-mentioned image feature map group to 0, and to set the feature values ​​outside the target detection area in the non-infrared image feature map included in the above-mentioned image feature map group to 0, to obtain an updated image feature map group.

[0048] The updated image feature map group includes an updated infrared image feature map and an updated non-infrared image feature map. The target detection region is the candidate detection region with the maximum regional confidence in the candidate detection region set. The candidate detection region with the maximum regional confidence should include the complete target object and the area to be detected. In practice, since body posture detection is only required for target objects within the target detection region, to reduce the amount of subsequent feature processing, the feature values ​​outside the target detection region in the non-infrared image feature map and the infrared image feature map are set to 0, thereby reducing the amount of subsequent feature processing.

[0049] In the second step, for each updated image feature map group in the obtained updated image feature map group sequence, the updated infrared image feature map and the updated non-infrared image feature map included in the updated image feature map group are superimposed to obtain a superimposed image feature map.

[0050] In practice, since the feature map sizes of the updated infrared image feature map and the updated non-infrared image feature map are consistent, the feature maps can be directly superimposed to obtain the superimposed image feature map.

[0051] In the third step, a low-level image feature extractor is used to extract low-level image features from the obtained superimposed image feature map sequence to obtain a low-level image feature map sequence.

[0052] In practice, see Figure 6 Figure 1 shows a schematic diagram of the model structure of the body posture information prediction model, where the low-level image feature extractor includes a bottleneck layer, residual block A, residual block B, residual block C, residual block D, and residual block E. Specifically, the features output by residual block A and residual block B are fused using channel fusion, and the fused features serve as inputs to residual block C, residual block D, and residual block E, respectively. For example, if the feature size of residual block A is a×b×c1 and the feature size of residual block B is a×b×c2, then the feature size of the fused features obtained by channel fusion of the features output by residual block A and residual block B is a×b×c2.

[0053] The fourth step is to perform high-level image feature extraction on the above low-level image feature map sequence through a high-level image feature extractor to obtain a high-level image feature map sequence.

[0054] In practice, see further Figure 6 Figure 1 is a schematic diagram of the model structure of the body posture information prediction model, where the advanced image feature extractor includes: a multi-head attention mechanism module, a multi-layer perceptron and a 2D convolutional layer.

[0055] The fifth step is to generate the above-mentioned body posture information based on the above-mentioned high-level image feature map sequence and the body posture information predictor.

[0056] Among them, the above-mentioned dual-path image feature extraction model, the above-mentioned low-level image feature extractor, the above-mentioned high-level image feature extractor and the above-mentioned body posture information predictor are included in the body posture information prediction model, and the above-mentioned body posture information predictor includes: a key point predictor, a key point type classifier and a posture type classifier.

[0057] In practice, see further Figure 6 The following figure shows the model structure of the keypoint predictor body posture information prediction model. The input of the 2D convolutional layer is fed into the keypoint predictor, keypoint type classifier, and posture type classifier in parallel. Specifically, the keypoint predictor can use a heatmap regression predictor. The keypoint type classifier can be implemented using multiple classifiers to classify the body organ type at the keypoint, and the posture type classifier can be implemented using multiple classifiers to classify the posture type of the target object.

[0058] Optionally, the above method further includes:

[0059] In response to the body posture information indicating that the target object has an abnormal posture, a reminder for posture adjustment of the target object is initiated through a reminder device included in the nursing bed, or a reminder for abnormal posture of the target object is initiated to a remote monitoring terminal.

[0060] In some embodiments, the above-mentioned execution subject can respond to the body posture information indicating that the target object has an abnormal posture, and initiate a posture adjustment reminder for the target object through the reminder device included in the nursing bed, or initiate a posture abnormality reminder for the target object to a remote monitoring terminal. In practice, in order to avoid the accumulation of patient wounds and thus affect recovery, a body posture type that is not recommended can be set for the target object. When the posture type included in the body posture information is consistent with the body posture type that is not recommended, the target object's posture can be considered abnormal. The reminder device can be a vibration reminder device installed at the head of the nursing bed. In practice, since the nursing ward contains at least one nursing bed, in order to avoid the sound emitted by the reminder device affecting other patients, a vibration reminder can be used to remind the target object to adjust its posture. The remote monitoring terminal can be a monitoring terminal installed at a nurse's station for monitoring multiple nursing wards. By sending a posture abnormality reminder to the remote monitoring terminal, a nurse or doctor is proactively reminded to check the current status of the target object.

[0061] Step 103 : In response to the body posture information indicating that the target object has a normal posture and the posture type is a lying posture, a real-time temperature signal and a real-time humidity signal are collected through a temperature and humidity sensor included in the nursing bed.

[0062] In some embodiments, the above-mentioned execution subject can respond to the body posture information to characterize that the posture of the target object is normal, and the posture type is a lying type, and collect real-time temperature signals and real-time humidity signals through the temperature and humidity sensors included in the nursing bed. In practice, the nursing bed includes: a temperature and humidity signal collection area. Among them, the above-mentioned temperature and humidity signal collection area corresponds to the waist and hip position of the target object in the standard lying position, and the above-mentioned temperature and humidity sensors are arranged in an array within the above-mentioned temperature and humidity signal collection area. Therefore, at each moment, the matrix temperature and humidity signal values ​​can be collected, and the temperature and humidity signal values ​​at multiple moments constitute the real-time temperature signal and real-time humidity signal. Specifically, the temperature and humidity sensors are arranged under the cushion layer of the nursing bed to prevent the target object from having a foreign body sensation when lying down.

[0063] Step 104: Determine temperature and humidity control information based on the real-time temperature signal and the real-time humidity signal.

[0064] In some embodiments, the execution entity may determine temperature and humidity control information based on the real-time temperature signal and the real-time humidity signal. The temperature and humidity control information includes a control mode, a control area, and a control level. In practice, control modes include ventilation mode and heating mode. The control level may represent the degree of control under different control modes. For example, using the heating mode as an example, the control levels may include heating level, constant temperature level, and sleep level.

[0065] In some optional implementations of some embodiments, the execution subject determines the temperature and humidity control information according to the real-time temperature signal and the real-time humidity signal, which may include the following steps:

[0066] In the first step, signal features are extracted from the real-time temperature signal and the real-time humidity signal to generate a temperature feature graph sequence and a humidity feature graph sequence.

[0067] In practice, because temperature and humidity sensors are arranged in an array, the humidity and temperature values ​​at a single moment in time for real-time temperature and humidity signals are represented in matrix form. Therefore, an RNN (Recurrent Neural Network) model can be used to extract signal features from these real-time temperature and humidity signals to generate sequences of temperature and humidity feature graphs.

[0068] The second step is to intercept the updated infrared image feature map included in the updated image feature map group sequence and the local infrared image feature map at the position corresponding to the temperature and humidity signal acquisition area for each updated image feature map group sequence in the updated image feature map group sequence.

[0069] In practice, since the temperature and humidity signal acquisition area is located in the area to be detected, and considering that the infrared image can characterize the temperature changes in the area to be detected, it can be used as one of the bases for temperature control. In order to avoid the waste of computing resources caused by repeated feature extraction, the updated infrared image feature map included in the updated image feature map group can be directly intercepted, and the local infrared image feature map at the corresponding position of the above-mentioned temperature and humidity signal acquisition area can be directly intercepted.

[0070] The third step is to generate the above-mentioned temperature and humidity control information based on the obtained local infrared image feature map sequence, temperature feature map sequence and humidity feature map sequence and the pre-trained temperature and humidity control information prediction model.

[0071] The temperature and humidity control information prediction model includes a deep feature extraction model, a control mode classifier, a control area predictor, and a control gear classifier. In practice, the deep feature extraction model can be three convolutional neural network models arranged in parallel. Specifically, the three convolutional neural network models are connected to a feature fusion layer for fusing the outputs of the three convolutional neural network models. The control area predictor can be implemented using a coordinate regressor to predict the coordinates of the control area. Both the control mode classifier and the control gear classifier can be implemented using multiple classifiers.

[0072] Step 105: Control the temperature and humidity adjustment device included in the nursing bed to adjust the temperature and humidity according to the temperature and humidity control information.

[0073] In some embodiments, the above-mentioned execution entity can control the temperature and humidity adjustment device included in the nursing bed to adjust the temperature and humidity according to the temperature and humidity control information.

[0074] Optionally, the temperature and humidity control device includes: a ventilation device and a heating device. The ventilation device is arranged on the back of the nursing bed. The ventilation device is composed of an array of cooling fans. The cooling fans included in the ventilation device are independently controlled. The heating device is arranged on the front of the nursing bed. The heating device is composed of a matrix of heating cloth. The heating cloth included in the heating device is independently controlled. To avoid direct contact between the heating cloth and the skin, thereby causing problems such as low-temperature burns, a thermally conductive lining cloth woven from polyester yarn is provided on the upper layer of the heating cloth. The heating cloth is woven from flame-retardant material and has a heating wire embedded therein. The heating wire is electrically heated by a heating circuit. Since the cushioning layers of the nursing bed and the bedding are ventilated structures, the cooling fans included in the ventilation device can adopt an air suction method, that is, the wind direction flows from the upper side of the nursing bed to the lower side of the nursing bed, to achieve a heat dissipation effect on the target object on the nursing bed. The array design can be independently controlled, thereby achieving a more precise temperature control and ventilation effect.

[0075] In some optional implementations of some embodiments, the execution subject controls the temperature and humidity adjustment device included in the nursing bed to adjust the temperature and humidity according to the temperature and humidity control information, including:

[0076] The first step is to activate the cooling fan in the above-mentioned ventilation device corresponding to the control area included in the above-mentioned temperature and humidity control information in response to the control mode included in the above-mentioned temperature and humidity control information being the temperature control mode, and to control the fan speed of the cooling fan in the above-mentioned ventilation device corresponding to the control area included in the above-mentioned temperature and humidity control information according to the control gear included in the above-mentioned temperature and humidity control information.

[0077] In practice, since the camera has been calibrated, that is, the coordinates between the geodetic coordinate system and the image coordinate system can be directly switched, when the control area is known, the fan identification of the cooling fan located in the control area can be obtained by coordinate mapping. Combined with the fact that the cooling fan can be controlled independently, only the cooling fan located in the control area can be activated for heat dissipation.

[0078] In the second step, in response to the control mode included in the above-mentioned temperature and humidity control information being the humidity control mode, the heating cloth in the above-mentioned heating device corresponding to the control area included in the above-mentioned temperature and humidity control information is activated, and according to the control gear included in the above-mentioned temperature and humidity control information, the heating temperature of the heating cloth in the above-mentioned heating device corresponding to the control area included in the above-mentioned temperature and humidity control information is controlled.

[0079] In practice, since the camera has been calibrated, that is, the coordinates between the geodetic coordinate system and the image coordinate system can be directly switched, when the control area is known, the heating cloth identification of the heating cloth located in the control area can be obtained by coordinate mapping. Combined with the fact that the heating cloth can be independently controlled, only the heating cloth located in the control area can be activated for heating.

[0080] Step 106 : In response to obtaining the active control signal, performing signal analysis on the active control signal to generate nursing bed control information.

[0081] In some embodiments, the execution entity may, in response to receiving an active control signal, perform signal analysis on the active control signal to generate nursing bed control information. The active control signal is a control signal initiated by a signal source, which may include a remote monitoring terminal or a nursing bed controller. In practice, steps 101 to 106 provide passive nursing bed control, particularly for patients without conscious awareness or the ability to move independently, allowing for effective monitoring. Furthermore, for patients with conscious awareness or the ability to move independently, or for monitoring and control initiated remotely by nurses and doctors, active control in step 108 may also be employed to control the nursing bed. Specifically, the active control signal may be initiated by the remote monitoring terminal via a wired or wireless connection, or by a wired nursing bed controller on the nursing bed side. Upon receipt of the active control signal, the active control signal may be analyzed to obtain control instructions as nursing bed control information. In practice, nursing bed control information includes control mode, control gear, and control area.

[0082] In some optional implementations of some embodiments, the execution subject, in response to obtaining the active control signal, performs signal analysis on the active control signal to generate nursing bed control information, including:

[0083] The first step is to perform content recognition on the voice signal in response to the active control signal being a voice signal to generate the nursing bed control information.

[0084] In practice, for patients with conscious but limited mobility, the nursing bed controller can receive active control signals through voice input. In this case, voice recognition of the active control signals is required to generate the nursing bed control information. Specifically, models such as recurrent neural networks and long short-term memory networks can be used to perform content recognition on the voice signals to generate the nursing bed control information.

[0085] The second step is to respond to the above-mentioned active control signal non-voice signal, perform instruction analysis on the above-mentioned active control signal to obtain the above-mentioned nursing bed control information.

[0086] In practice, when the active control signal is a non-voice signal, the nursing bed control information can be obtained through function mapping. For example, when the user presses the heating mode button and gear selection button included in the nursing bed controller, the corresponding active control signal can control the heating cloth to heat according to the gear.

[0087] Step 107: Control the nursing bed according to the nursing bed control information.

[0088] In some embodiments, the aforementioned execution entity can control the nursing bed based on the nursing bed control information. In practice, because the nursing bed control information includes a control mode, a control level, and a control area, the cooling fan or heating cloth corresponding to the control area can be activated based on the control mode, and heat or heat can be provided based on the control level.

[0089] For example, see Figure 7 The control effects of the nursing bed in heating mode and ventilation mode are shown in the schematic diagram. The nursing bed can be controlled by the nursing bed controller 701. Specifically, the nursing bed controller may include a control mode knob, a control gear knob, and a control area selection screen. The control area selection screen can be touch-sensitive to select the control area. In heating mode, the heating device 702 can be controlled to heat the heating cloth within the corresponding control area. In cooling mode, the cooling device 703 can be controlled to cool the cooling fan within the corresponding control area.

[0090] Step 108 : In response to reaching the predetermined time point, a body cleaning reminder for the target object is initiated to the remote monitoring terminal.

[0091] The nursing bed also includes an auxiliary cleaning device for the patient. In practice, the auxiliary cleaning device can be a buttocks cleaning device arranged under the nursing bed. Specifically, the buttocks cleaning device can be a washing basin. The washing basin is arranged under the nursing bed in a pull-out manner. The patient's buttocks can be cleaned regularly by setting up a washing basin. In addition, for patients who have no autonomous consciousness or no autonomous movement ability, there may be a problem of excrement not being able to be excreted independently. At the same time, taking into account the uneven doctor-patient ratio, doctors or nurses are reminded to clean the target patient's body through regular reminders to avoid the problem of the target patient excreting on the nursing bed.

[0092] In some optional implementations of some embodiments, the above method further includes:

[0093] The first step is to respond to the remote monitoring terminal initiating a remote monitoring instruction for the nursing bed, for each real-time image group in the real-time image group sequence, perform Gaussian blur on the image content of the real-time image group outside the corresponding target detection area to obtain a blurred real-time image group.

[0094] In practice, the remote monitoring terminal can also be a mobile terminal used by the family member of the target patient. This allows for remote monitoring by the family member. Furthermore, considering the privacy of other patients, during video transmission, the real-time image group must be Gaussian blurred outside the corresponding target detection area.

[0095] In the second step, the blurred real-time image group sequence is end-to-end encrypted to obtain the encrypted video.

[0096] In practice, end-to-end encryption is used to ensure the security of transmitted video transmission.

[0097] The third step is to send the encrypted video to the remote monitoring terminal.

[0098] The above-described embodiments of the present disclosure have the following beneficial effects: Through the control and cleaning methods applied to nursing beds according to some embodiments of the present disclosure, effective and timely monitoring of patient posture is achieved, while effectively addressing the problem of body temperature imbalance caused by long-term bed rest. Specifically, the present disclosure first captures a real-time image sequence in response to a target subject being located within a detection area. The detection area is the area surrounding the nursing bed, and the target subject is wearing a wireless radio frequency bracelet associated with the bed. The real-time image sequence includes real-time infrared images and real-time non-infrared images. In practice, the real-time image sequence is captured only when the target subject (patient) is within the detection area. This avoids capturing images when the target subject is not within the detection area, reducing the number of images captured and the subsequent image processing required. Secondly, based on the real-time image sequence, body posture information corresponding to the target subject is determined. The body posture information includes a set of torso key point information and a posture type. In practice, since the target subject may be covered by objects (e.g., bedding) when lying on the nursing bed, body posture recognition is performed by combining infrared and non-infrared images. Furthermore, in response to the body posture information indicating that the target subject is in a normal posture and the posture type is a reclining posture, a real-time temperature signal and a real-time humidity signal are collected via a temperature and humidity sensor included in the nursing bed. Furthermore, temperature and humidity control information is determined based on the real-time temperature and humidity signals, wherein the temperature and humidity control information includes a control mode, a control area, and a control gear. Then, based on the temperature and humidity control information, a temperature and humidity adjustment device included in the nursing bed is controlled to adjust the temperature and humidity. By combining the temperature and humidity sensor, whether the target subject experiences body temperature imbalance while in a reclining position on the nursing bed is analyzed, and the temperature and humidity are adjusted accordingly. Furthermore, in response to obtaining an active control signal, the active control signal is analyzed to generate nursing bed control information. Then, the nursing bed is controlled based on the nursing bed control information. Finally, in response to reaching a predetermined time point, a body cleaning reminder for the target subject is sent to a remote monitoring terminal, wherein the nursing bed also includes an auxiliary cleaning device for the patient. This approach effectively addresses the issue of medical staff being unable to monitor patients’ posture effectively and promptly due to the imbalance in the doctor-patient ratio. It also effectively addresses issues such as body temperature imbalances caused by prolonged bed rest.

[0099] Further references Figure 8 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a control and cleaning device for a nursing bed. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the control and cleaning device applied to the nursing bed can be specifically applied to various electronic devices.

[0100] like Figure 8 As shown, some embodiments of the control and cleaning device 800 applied to a nursing bed include: a first acquisition unit 801, a first determination unit 802, a second acquisition unit 803, a second determination unit 804, a first control unit 805, a signal analysis unit 806, a second control unit 807, and a body cleaning reminder unit 808. The first acquisition unit 801 is configured to acquire a real-time image group sequence in response to a target object being located in a to-be-detected area, wherein the to-be-detected area is the area where the nursing bed is located, the target object wears a wireless radio frequency bracelet bound to the identity of the nursing bed, and the real-time image group includes: a real-time infrared image and a real-time non-infrared image; the first determination unit 802 is configured to determine the body posture information corresponding to the target object based on the real-time image group sequence, wherein the body posture information includes: a torso key point information set and a posture type; the second acquisition unit 803 is configured to acquire a real-time temperature signal and a real-time humidity signal through the temperature and humidity sensor included in the nursing bed in response to the body posture information indicating that the target object's posture is normal and the posture type is a lying posture; the second determination unit 804 is configured to acquire a real-time temperature signal and a real-time humidity signal through the temperature and humidity sensor included in the nursing bed in response to the body posture information indicating that the target object's posture is normal and the posture type is a lying posture; It is configured to determine the temperature and humidity control information according to the above-mentioned real-time temperature signal and the above-mentioned real-time humidity signal, wherein the above-mentioned temperature and humidity control information includes: control mode, control area and control gear; the first control unit 805 is configured to control the temperature and humidity adjustment device included in the above-mentioned nursing bed to adjust the temperature and humidity according to the above-mentioned temperature and humidity control information; the signal analysis unit 806 is configured to perform signal analysis on the above-mentioned active control signal in response to obtaining the active control signal to generate nursing bed control information; the second control unit 807 is configured to control the above-mentioned nursing bed according to the above-mentioned nursing bed control information; the body cleaning reminder unit 808 is configured to initiate a body cleaning reminder for the above-mentioned target object to the remote monitoring terminal in response to reaching a predetermined time point, wherein the nursing bed also includes an auxiliary cleaning device for the patient.

[0101] It is understood that the various units described in the control and cleaning device 800 for the nursing bed are similar to those described in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the control and cleaning device 800 for the nursing bed and the units contained therein, and will not be repeated here.

[0102] Reference below Figure 9 , which shows a structural schematic diagram of an electronic device (eg, a computing device) 900 suitable for implementing some embodiments of the present disclosure. Figure 9The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0103] like Figure 9 As shown, electronic device 900 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 901, which can perform various appropriate actions and processes based on programs stored in read-only memory 902 or programs loaded from storage device 908 into random access memory 903. Random access memory 903 also stores various programs and data required for the operation of electronic device 900. Processing device 901, read-only memory 902, and random access memory 903 are connected to each other via bus 904. Input / output interface 905 is also connected to bus 904.

[0104] Typically, the following devices may be connected to the I / O interface 905: an input device 906 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 907 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 908 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 909. The communication device 909 may allow the electronic device 900 to communicate with other devices wirelessly or by wire to exchange data. Figure 9 The electronic device 900 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead. Figure 9 Each block shown in the figure may represent one device, or may represent multiple devices as needed.

[0105] In particular, according to some embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from a network via the communication device 909, or installed from the storage device 908, or installed from the read-only memory 902. When the computer program is executed by the processing device 901, the above-mentioned functions defined in the method of some embodiments of the present disclosure are performed.

[0106] It should be noted that the computer-readable medium described in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable storage media may include, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or components, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or component. Furthermore, in some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. This propagated data signal may take a variety of forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wire, optical cable, RF (radio frequency), or any suitable combination thereof.

[0107] In some embodiments, the client and server can communicate using any currently known or later developed network protocol, such as HTTP (Hypertext Transfer Protocol), and can be interconnected with 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"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or later developed network.

[0108] The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or it may exist independently without being assembled into the electronic device. The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: in response to the target object being located in the area to be detected, collects a real-time image group sequence, wherein the above-mentioned area to be detected is the area where the nursing bed is located, and the above-mentioned target object wears a wireless radio frequency bracelet bound to the identity of the above-mentioned nursing bed, and the real-time image group includes: real-time infrared image and real-time non-infrared image; according to the above-mentioned real-time image group sequence, determines the body posture information corresponding to the above-mentioned target object, wherein the above-mentioned body posture information includes: a torso key point information set and a posture type; in response to the above-mentioned body posture information indicating that the posture of the above-mentioned target object is normal, and the above-mentioned posture type is a lying type, through the above-mentioned nursing The bed includes a temperature and humidity sensor that collects real-time temperature signals and real-time humidity signals; determines temperature and humidity control information based on the above-mentioned real-time temperature signals and the above-mentioned real-time humidity signals, wherein the above-mentioned temperature and humidity control information includes: control mode, control area and control gear; controls the temperature and humidity adjustment device included in the above-mentioned nursing bed to adjust the temperature and humidity based on the above-mentioned temperature and humidity control information; in response to obtaining an active control signal, performs signal analysis on the above-mentioned active control signal to generate nursing bed control information; controls the above-mentioned nursing bed based on the above-mentioned nursing bed control information; in response to reaching a predetermined time point, initiates a body cleaning reminder for the above-mentioned target object to the remote monitoring terminal, wherein the nursing bed also includes an auxiliary cleaning device for the patient.

[0109] Computer program code for performing the operations of some embodiments of the present disclosure may be written 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 through 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., through the Internet using an Internet service provider).

[0110] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0111] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The units described may also be provided in a processor. For example, they may be described as follows: a processor comprising a first acquisition unit, a first determination unit, a reminder or prompt unit, a second acquisition unit, a second determination unit, a first control unit, a signal parsing unit, and a second control unit. The names of these units do not, in some cases, constitute limitations on the units themselves. For example, the second control unit may also be described as a "unit for controlling the nursing bed according to the nursing bed control information."

[0112] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0113] The above descriptions are merely some preferred embodiments of the present disclosure and illustrate the underlying technical principles. Those skilled in the art should understand that the scope of the invention encompassed by the embodiments of the present disclosure is not limited to technical solutions formed by specific combinations of the aforementioned technical features. It also encompasses other technical solutions formed by any combination of the aforementioned technical features or their equivalents, without departing from the aforementioned inventive concept. For example, a technical solution formed by replacing the aforementioned features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A control and cleaning method for a nursing bed, comprising: In response to a target object being located within an area to be detected, a sequence of real-time image groups is collected, wherein the area to be detected is an area where a nursing bed is located, and the target object is wearing a wireless radio frequency bracelet that is bound to the identity of the nursing bed. The real-time image group includes: a real-time infrared image and a real-time non-infrared image collected by a camera, and the camera also includes a radio frequency receiver that is communicatively connected to a radio frequency transmitter contained in the wireless radio frequency bracelet for locating the target object; Determine body posture information corresponding to the target object based on the real-time image group sequence, wherein the body posture information includes: a torso key point information set and a posture type. For each real-time image group in the real-time image group sequence, perform the following steps: Performing image feature extraction on the real-time infrared image and the real-time non-infrared image included in the real-time image group using a dual-path image feature extraction model to generate an image feature map group including infrared image feature maps and non-infrared image feature maps with consistent feature map sizes; Constructing a basic detection area according to the area to be detected, wherein the area corner points of the area to be detected are located on the area boundary of the basic detection area; Constructing a set of candidate detection areas according to the basic detection area and a preset window magnification ratio, wherein the basic detection area is a region subset of the candidate detection areas in the set of candidate detection areas; For each candidate detection area in the set of candidate detection areas, determining a region confidence of the candidate detection area based on the candidate detection area, the infrared image feature map, and the non-infrared image feature map, wherein the region confidence represents a confidence level of whether the candidate detection area contains a target object; Setting the feature values ​​outside the target detection area in the infrared image feature map included in the image feature map group to 0, and setting the feature values ​​outside the target detection area in the non-infrared image feature map included in the image feature map group to 0, to obtain updated infrared image feature maps and non-infrared image feature maps, where the target detection area is the candidate detection area corresponding to the maximum area confidence in the candidate detection area set; generating body posture information based on the updated infrared image feature map and the non-infrared image feature map; In response to the body posture information indicating that the target object has a normal posture and the posture type is a lying posture, collecting a real-time temperature signal and a real-time humidity signal through a temperature and humidity sensor included in the nursing bed; Determining temperature and humidity control information according to the real-time temperature signal and the real-time humidity signal, wherein the temperature and humidity control information includes: a control mode, a control area, and a control gear; Controlling the temperature and humidity adjustment device included in the nursing bed to adjust the temperature and humidity according to the temperature and humidity control information; In response to acquiring the active control signal, performing signal analysis on the active control signal to generate nursing bed control information; Controlling the nursing bed according to the nursing bed control information; In response to reaching a predetermined time point, a body cleaning reminder for the target object is initiated to the remote monitoring terminal, wherein the nursing bed also includes an auxiliary cleaning device for the patient.

2. The method according to claim 1, wherein The method further comprises: In response to the body posture information indicating that the target object has an abnormal posture, a reminder for posture adjustment of the target object is initiated through a reminder device included in the nursing bed, or a reminder for abnormal posture of the target object is initiated to a remote monitoring terminal.

3. The method according to claim 1, wherein In response to acquiring the active control signal, performing signal parsing on the active control signal to generate nursing bed control information includes: In response to the active control signal being a voice signal, performing content recognition on the voice signal to generate the nursing bed control information; In response to the active control signal being a non-voice signal, the active control signal is subjected to instruction parsing to obtain the nursing bed control information.

4. The method according to claim 3, wherein: The temperature and humidity regulating device includes: a ventilation device and a heating device, wherein the ventilation device is arranged at the back of the nursing bed, the ventilation device is composed of an array of cooling fans, and the cooling fans included in the ventilation device are independently controlled; the heating device is arranged at the front of the nursing bed, the heating device is composed of a matrix of heating cloths, and the heating cloths included in the heating device are independently controlled; and The step of controlling the temperature and humidity adjustment device included in the nursing bed to adjust the temperature and humidity according to the temperature and humidity control information includes: In response to the control mode included in the temperature and humidity control information being the temperature control mode, activating a cooling fan in the ventilation device corresponding to the control area included in the temperature and humidity control information, and controlling a fan speed of the cooling fan in the ventilation device corresponding to the control area included in the temperature and humidity control information according to a control gear included in the temperature and humidity control information; In response to the control mode included in the temperature and humidity control information being a humidity control mode, the heating cloth in the heating device corresponding to the control area included in the temperature and humidity control information is activated, and the heating temperature of the heating cloth in the heating device corresponding to the control area included in the temperature and humidity control information is controlled according to the control gear included in the temperature and humidity control information.

5. The method according to claim 4, wherein The step of determining body posture information corresponding to the target object according to the real-time image group sequence further includes: For each updated image feature map group in the obtained updated image feature map group sequence, superimposing the updated infrared image feature map and the updated non-infrared image feature map included in the updated image feature map group to obtain a superimposed image feature map; Performing low-level image feature extraction on the obtained superimposed image feature map sequence through a low-level image feature extractor to obtain a low-level image feature map sequence; Performing high-level image feature extraction on the low-level image feature map sequence using a high-level image feature extractor to obtain a high-level image feature map sequence; The body posture information is generated according to the high-level image feature map sequence and the body posture information predictor, wherein the two-way image feature extraction model, the low-level image feature extractor, the high-level image feature extractor and the body posture information predictor are included in a body posture information prediction model, and the body posture information predictor includes: a key point predictor, a key point type classifier and a posture type classifier.

6. The method according to claim 5, wherein: The nursing bed comprises: a temperature and humidity signal acquisition area, wherein the temperature and humidity signal acquisition area corresponds to the waist and hip position of the target subject in a standard lying position, and the temperature and humidity sensors are arranged in an array in the temperature and humidity signal acquisition area; and The determining of temperature and humidity control information according to the real-time temperature signal and the real-time humidity signal includes: Performing signal feature extraction on the real-time temperature signal and the real-time humidity signal to generate a temperature feature graph sequence and a humidity feature graph sequence; For each updated image feature map group sequence in the updated image feature map group sequence, intercepting the updated infrared image feature map included in the updated image feature map group and the local infrared image feature map at a position corresponding to the temperature and humidity signal acquisition area; The temperature and humidity control information is generated based on the obtained local infrared image feature map sequence, temperature feature map sequence and humidity feature map sequence and a pre-trained temperature and humidity control information prediction model, wherein the temperature and humidity control information prediction model includes: a deep feature extraction model, a control mode classifier, a control area predictor and a control gear classifier.

7. The method according to claim 6, wherein: The method further comprises: In response to the remote monitoring terminal initiating a remote monitoring instruction for the nursing bed, for each real-time image group in the real-time image group sequence, performing Gaussian blurring on image content of the real-time image group outside a corresponding target detection area to obtain a blurred real-time image group; End-to-end encryption is performed on the blurred real-time image group sequence to obtain an encrypted video; The encrypted video is sent to the remote monitoring terminal.

8. A control and cleaning device for a nursing bed, comprising: A first acquisition unit is configured to acquire a sequence of real-time image groups in response to a target object being located within an area to be detected, wherein the area to be detected is an area where a nursing bed is located, and the target object is wearing a wireless radio frequency bracelet that is bound to the identity of the nursing bed, and the real-time image group includes: a real-time infrared image and a real-time non-infrared image acquired by a camera, and the camera also includes a radio frequency receiver that is communicatively connected to a radio frequency transmitter included in the wireless radio frequency bracelet for locating the target object; The first determining unit is configured to determine body posture information corresponding to the target object based on the real-time image group sequence, wherein the body posture information includes: a torso key point information set and a posture type, and for each real-time image group in the real-time image group sequence, perform the following steps: Performing image feature extraction on the real-time infrared image and the real-time non-infrared image included in the real-time image group using a dual-path image feature extraction model to generate an image feature map group including infrared image feature maps and non-infrared image feature maps with consistent feature map sizes; Constructing a basic detection area according to the area to be detected, wherein the area corner points of the area to be detected are located on the area boundary of the basic detection area; Constructing a set of candidate detection areas according to the basic detection area and a preset window magnification ratio, wherein the basic detection area is a region subset of the candidate detection areas in the set of candidate detection areas; For each candidate detection area in the set of candidate detection areas, determining a region confidence of the candidate detection area based on the candidate detection area, the infrared image feature map, and the non-infrared image feature map, wherein the region confidence represents a confidence level of whether the candidate detection area contains a target object; Setting the feature values ​​outside the target detection area in the infrared image feature map included in the image feature map group to 0, and setting the feature values ​​outside the target detection area in the non-infrared image feature map included in the image feature map group to 0, to obtain updated infrared image feature maps and non-infrared image feature maps, where the target detection area is the candidate detection area corresponding to the maximum area confidence in the candidate detection area set; generating body posture information based on the updated infrared image feature map and the non-infrared image feature map; a second collecting unit configured to collect a real-time temperature signal and a real-time humidity signal via a temperature and humidity sensor included in the nursing bed in response to the body posture information indicating that the posture of the target object is normal and the posture type is a lying type; a second determining unit configured to determine temperature and humidity control information according to the real-time temperature signal and the real-time humidity signal, wherein the temperature and humidity control information includes: a control mode, a control area, and a control gear; a first control unit configured to control the temperature and humidity adjustment device included in the nursing bed to adjust the temperature and humidity according to the temperature and humidity control information; a signal parsing unit configured to, in response to acquiring the active control signal, perform signal parsing on the active control signal to generate nursing bed control information; a second control unit configured to control the nursing bed according to the nursing bed control information; The body cleaning reminder unit is configured to initiate a body cleaning reminder for the target object to the remote monitoring terminal in response to reaching a predetermined time point, wherein the nursing bed also includes an auxiliary cleaning device for the patient.

9. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

Images