Nursing service testing methods, devices, electronic equipment, storage media and products

Abstract

This invention discloses a method, device, electronic device, storage medium, and product for detecting nursing services. The method includes: generating first information corresponding to multiple first objects within a first region; generating second information corresponding to the multiple first objects, wherein the second information indicates the role identification result of the first objects, and the second information is related to the spatiotemporal behavioral characteristics of the first objects, the spatiotemporal behavioral characteristics of the first objects originating from the first information of the first objects; identifying third information of a second object within a second region, wherein the second object is the role among the multiple first objects used to provide nursing services to a recipient, and the third information indicates the spatiotemporal micro-motion characteristics generated by the second object in the second region; and detecting whether the nursing services in the first region are compliant based on the third information and the second information. This application's solution achieves compliance checks on nursing services.

Description

Technical Field

[0001] This invention relates to the field of radar technology, and more particularly to a method, apparatus, electronic device, storage medium, and product for detecting nursing services. Background Technology

[0002] In the actual implementation of nursing services, it is necessary to strictly verify the presence of caregivers, the condition of the recipients, the authenticity of nursing actions, and whether the duration and actions of nursing care comply with regulations, in order to safeguard the long-term care insurance fund and ensure the quality of nursing services. Currently, compliance checks of nursing services mainly rely on manual spot checks, paper records, video playback, and location check-in. These methods suffer from high costs, poor real-time performance, difficulty in continuously, objectively, and automatically verifying each nursing process, and high costs and difficulties in retrospective investigation. Ordinary check-in or location check-in can only confirm the presence of caregivers, but cannot prove that substantive nursing actions have occurred, which can easily lead to situations such as presence without providing care or leaving after a short stay. In addition, video surveillance solutions pose a risk of privacy leaks and are limited in deployment in sensitive scenarios such as bedridden care and toilet assistance, making them unsuitable. Summary of the Invention

[0003] This invention provides a nursing service detection method, device, electronic device, storage medium, and program product, which can verify whether nursing staff are present, whether the person being cared for is in a caring state, whether the nursing service actually occurs, and whether the nursing process meets the prescribed duration and prescribed action requirements.

[0004] According to one aspect of the present invention, a method for detecting nursing services is provided, the method comprising: Generate first information corresponding to multiple first objects within a first region, wherein the first information indicates the trajectory, spatial position, and velocity of the first object; Generate second information corresponding to the plurality of first objects. The second information indicates the role recognition result of the first object. The second information is related to the spatiotemporal behavioral characteristics of the first object. The spatiotemporal behavioral characteristics of the first object are derived from the first information of the first object. The third information of the second object in the second region is identified. The second object is the object role among the plurality of first objects used to provide nursing services to the object being cared for. The third information indicates the spatiotemporal micro-motion characteristics of the second object in the second region. The second region is a local area in the first region used to characterize the second object performing nursing actions around the object being cared for. Based on the third information and the second information, the compliance of the nursing services in the first area is detected.

[0005] According to another aspect of the present invention, a nursing service detection device is provided, the device comprising: The generation module is used to generate first information corresponding to multiple first objects within a first region, wherein the first information indicates the trajectory, spatial position, and velocity of the first object. The generation module is also used to generate second information corresponding to the plurality of first objects. The second information indicates the role recognition result of the first object. The second information is related to the spatiotemporal behavior characteristics of the first object. The spatiotemporal behavior characteristics of the first object are derived from the first information of the first object. The identification module is used to identify third information of a second object in a second region. The second object is an object role among the plurality of first objects that provides nursing services to the object being cared for. The third information indicates the spatiotemporal micro-motion characteristics of the second object in the second region. The second region is a local area in the first region that represents the second object performing nursing actions around the object being cared for. The detection module is used to detect whether the nursing services in the first area are compliant based on the third information and the second information.

[0006] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the nursing service detection method according to any embodiment of the present invention.

[0007] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the nursing service detection method according to any embodiment of the present invention.

[0008] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements the nursing service detection method according to any embodiment of the present invention.

[0009] The technical solution of this invention generates first information about multiple first objects within a first region. This first information includes the trajectory, spatial location, and velocity of the first objects, enabling a quantitative representation of the motion state of the first objects within the first region. Based on this first information, spatiotemporal behavioral features of the first objects are extracted. Then, based on these spatiotemporal behavioral features, a role recognition result strongly correlated with the behavioral patterns of the first objects is generated. Quantifiable spatiotemporal behavioral features serve as the basis for role classification, accurately distinguishing the roles of each first object within the first region and providing a basis for screening nursing service-related objects. A second region representing the care space is delineated from the first region, and second objects belonging to nursing service roles are identified. These second objects are specifically extracted for providing nursing services to the recipients. The second object in the care service exhibits spatiotemporal micro-motion characteristics within the second area, enabling layered perception from macroscopic trajectories to refined micro-movements within the care area. This effectively filters out irrelevant walking behaviors and accurately captures the characteristics of real nursing operations, overcoming the limitation of relying solely on macroscopic movements to identify refined nursing actions. Furthermore, by integrating role recognition results with the spatiotemporal micro-motion characteristics of the second object in the second area for joint judgment, the system confirms that the object possesses the identity attributes of a nursing staff member and verifies the existence of micro-movement behaviors that conform to nursing characteristics within the care area. This significantly reduces the probability of misjudgments and omissions in nursing services, enabling objective, stable, and automated completion of nursing service compliance detection within the first area. This effectively improves the completeness, accuracy, and reliability of nursing service supervision in care scenarios.

[0010] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating a nursing service testing method provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating another nursing service testing method provided in an embodiment of the present invention; Figure 3 This is a flowchart illustrating another nursing service testing method provided in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the use of a role state machine in nursing service detection according to an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the generation of nursing events in nursing service detection according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of a nursing service detection device provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of an electronic device for implementing a nursing service detection method according to an embodiment of the present invention. Detailed Implementation

[0013] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0014] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0015] In some embodiments, Figure 1 This is a flowchart illustrating a nursing service testing method provided in an embodiment of the present invention. This embodiment is applicable to situations involving compliance checks of nursing services, particularly those related to long-term care or long-term nursing services. The method can be executed by a nursing service testing device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the nursing service detection method of this embodiment may include the following process: S110. Generate first information corresponding to multiple first objects within the first region. The first information indicates the trajectory, spatial position, and speed of the first object.

[0016] The first zone can be a monitoring space area specifically set up for home-based wards or elderly care residences. It can also be the entire boundary range for full-process monitoring and dynamic tracking of elderly care services. The first zone can be used to delineate the effective monitoring boundaries of care services, clarifying the coverage and control range of elderly care monitoring. The first object can be all objects continuously tracked by radar within the first zone. The first object can include potential care recipients, nursing staff providing care services to care recipients, and other personnel. Care recipients can refer to individuals with limited mobility, insufficient self-care ability, or whose health status requires continuous monitoring.

[0017] The first information may include the motion state information of the first object in the first region. This motion state information may include the motion trajectory, motion position, and motion speed. The motion trajectory may be a sequence of position coordinates of the first object within the first region over a continuous time period. It can be used to reconstruct the complete movement path of the first object and can be used to statistically analyze the first object's activity range, dwell time, movement trend, and frequency of entering and exiting the region. The motion position may be the position coordinates of the first object in the first region at each moment. The motion position can be used to determine the distance between different first objects within the first region and the location range of the first object. The motion speed may be the magnitude and direction of the speed calculated by differentiating the motion positions at adjacent moments.

[0018] Using the above scheme, all detected first objects can be continuously tracked within the first area, generating first information corresponding to each first object. The first information fully carries the motion trajectory, motion position, and motion speed formed in time sequence, realizing the continuous tracking and motion state quantification of all first objects, and providing data support for subsequent spatiotemporal behavior feature extraction, role recognition, and spatiotemporal micro-motion analysis.

[0019] S120. Generate second information corresponding to multiple first objects. The second information indicates the role recognition result of the first object. The second information is related to the spatiotemporal behavior characteristics of the first object. The spatiotemporal behavior characteristics of the first object are derived from the first information of the first object.

[0020] The role identification result of the first object can be the identity type to which the first object belongs. The role identification result can include the person being cared for, the nursing staff providing nursing services to the person being cared for, and personnel unrelated to the nursing services. The spatiotemporal behavioral characteristics of the first object can be the behavioral characteristics of the first object in the first area obtained by statistical analysis based on the movement trajectory, movement position, and movement speed indicated by the first information. The spatiotemporal behavioral characteristics of the first object can be used to indicate at least one of the following: the size of the activity range of the first object in the first area, the distribution pattern of the first object's residence in the first area, the rhythm of the first object's movement speed in the first area, the displacement change of the first object in the first area, and the degree of positional stability of the first object in the first area.

[0021] Using the above scheme, spatiotemporal behavioral features of each first object are extracted from the motion trajectory, position, and speed of the first object based on the first information. The role recognition of each first object in the first region is completed using the spatiotemporal behavioral features of the first object as input, and the role recognition result of each first object is generated. The role recognition of each first object can provide semantic support for subsequent spatiotemporal micro-motion analysis.

[0022] S130. Identify the third information of the second object in the second region. The second object is the object role among multiple first objects used to provide nursing services to the nursing recipient. The third information indicates the spatiotemporal micro-motion characteristics of the second object in the second region. The second region is a local area in the first region used to characterize the second object performing nursing actions around the nursing recipient.

[0023] The second zone can be a localized area within the first zone designated for performing caregiving actions related to nursing services. The second zone can be defined by beds, bedside areas, or care anchors within the first zone, which is the restricted area for caregivers to perform caregiving actions. The second zone can be used to filter out walking and passing-through actions unrelated to caregiving services.

[0024] The third piece of information can indicate the spatiotemporal micro-motion characteristics of the second object within the second area. These spatiotemporal micro-motion characteristics are small-scale, refined, and short-term quantitative representations of movement. These characteristics can indicate the caregiving actions performed by the caregiver providing nursing services to the recipient within the second area, distinguishing between ordinary walking and actual caregiving actions. The spatiotemporal micro-motion characteristics can be the localized, refined movement features of the second object within a defined spatial range and time window within the second area. These characteristics may include at least one of the following: the second object's dwell time percentage within the second area, average step length, step frequency, directional entropy, and positional drift. Furthermore, these characteristics can indicate high-frequency micro-movements, reciprocating movements, and multi-directional posture changes within the second area, filtering out actions unrelated to nursing services such as unidirectional passing and long-distance walking.

[0025] By adopting the above scheme, based on the role recognition results of the second information, second objects with nursing service roles are selected from all first objects. Then, a second area representing the bedside care space can be delineated within the first area. The spatiotemporal micro-movement features of the second objects within the second area are extracted to form third information, thereby realizing the quantitative characterization of the fine movements of nursing roles in the care area. By focusing only on the fine micro-movements of nursing staff in the care area, the interference of irrelevant personnel, irrelevant areas, and ordinary walking behavior can be eliminated, providing a basis for the compliance detection of nursing services.

[0026] S140. Based on the third and second information, detect whether the nursing services in the first area are compliant.

[0027] By integrating the role identification results of the second information with the spatiotemporal micro-motion characteristics of the second object in the second area indicated by the third information, compliance judgment and detection output are performed on care actions related to nursing services in the first area. This allows for the automatic identification of situations such as standardized on-duty care, missing care, false movement, and ineffective companionship, achieving intelligent compliance supervision of nursing services. Moreover, by integrating the dual dimensions of role identity semantics and refined spatiotemporal micro-motion characteristics, the system avoids the misjudgment problems caused by simple judgments based solely on location and distance. It can intelligently and automatically verify whether nursing staff are on duty, whether care services are actually being provided, and whether service behaviors comply with regulations, effectively improving the objectivity, automation level, and identification accuracy of long-term care scenario supervision.

[0028] The technical solution of this invention generates first information about multiple first objects within a first region. This first information includes the trajectory, spatial location, and velocity of the first objects, enabling a quantitative representation of the motion state of the first objects within the first region. Based on this first information, spatiotemporal behavioral features of the first objects are extracted. Then, based on these spatiotemporal behavioral features, a role recognition result strongly correlated with the behavioral patterns of the first objects is generated. Quantifiable spatiotemporal behavioral features serve as the basis for role classification, accurately distinguishing the roles of each first object within the first region and providing a basis for screening nursing service-related objects. A second region representing the care space is delineated from the first region, and second objects belonging to nursing service roles are identified. These second objects are specifically extracted for providing nursing services to the recipients. The second object in the care service exhibits spatiotemporal micro-motion characteristics within the second area, enabling layered perception from macroscopic trajectories to refined micro-movements within the care area. This effectively filters out irrelevant walking behaviors and accurately captures the characteristics of real nursing operations, overcoming the limitation of relying solely on macroscopic movements to identify refined nursing actions. Furthermore, by integrating role recognition results with the spatiotemporal micro-motion characteristics of the second object in the second area for joint judgment, the system confirms that the object possesses the identity attributes of a nursing staff member and verifies the existence of micro-movement behaviors that conform to nursing characteristics within the care area. This significantly reduces the probability of misjudgments and omissions in nursing services, enabling objective, stable, and automated completion of nursing service compliance detection within the first area. This effectively improves the completeness, accuracy, and reliability of nursing service supervision in care scenarios.

[0029] Figure 5 This is a flowchart illustrating another nursing service detection method provided by an embodiment of the present invention. The technical solution of this embodiment further optimizes the process of generating first information corresponding to multiple first objects within a first region, based on the technical solutions of the aforementioned embodiments. This embodiment can be combined with various optional solutions in one or more of the above embodiments. For example... Figure 5 As shown, the nursing service detection method of this embodiment may include the following process: S210. Generate a first trajectory of the first object in the first region. The first trajectory is related to the trajectory formed by the position of the first object in multiple first radar data. The multiple first radar data are radar data corresponding to multiple adjacent frames continuously output by the first radar in the first region. The first object in each adjacent frame of the multiple first radar data is successfully matched across frames and the number of successful cross-frame matching is greater than a preset number threshold.

[0030] The first radar data may include raw detection data output frame by frame by the first radar device. The first radar data may be the raw message stream output by a millimeter-wave radar scanning the first area. Multiple first radar data may be a sequence of temporally adjacent radar data frames continuously output by the first radar within the first area, providing temporally continuous raw detection data for cross-frame target association matching, avoiding target loss due to single-frame noise.

[0031] Cross-frame matching can be a process of associating and matching the first objects detected in radar data corresponding to adjacent frames, based on at least one of the following: positional distance, motion speed, and feature similarity between the first objects. This process allows for the assignment of a unique identifier to the same first object across different radar data frames, preventing object target identifier jumps and trajectory breaks. The number of successful cross-frame matchings can include the cumulative number of times the same first object has been identified as the same target in cross-frame matching across multiple consecutive adjacent frames. The preset threshold can be a pre-configured minimum number of valid matchings. This threshold serves as an admission condition; only when the number of consecutive successful cross-frame matchings exceeds the preset threshold is the target considered real and stable, allowing the generation of a valid first trajectory, while eliminating transient false detections and clutter targets.

[0032] The first trajectory can be a continuous motion trajectory of the first object in the current time period, formed by fitting the temporal position points of the first object in multiple consecutive frames of first radar data, provided that the cross-frame matching success count is met. The first trajectory can represent the movement path of the first object in the first region and can be used as a trajectory segment for trajectory association, position calculation, and velocity calculation.

[0033] In some embodiments, optionally, the first radar data may be the raw message stream output by a millimeter-wave radar scanning the first area within the first area. Generating the first trajectory of the first object in the first area includes: decoding the information byte sequence in the raw message stream included by multiple first radar data according to a predetermined protocol to obtain target list data in each frame of first radar data. Each target list data includes at least the identifier, position coordinates and velocity of the first object in the first radar data, and then the first trajectory of the first object in the first area can be generated based on the target list data in each frame of first radar data.

[0034] Using the above scheme, multiple frames of adjacent first radar data continuously output by the first radar can be acquired within the first area covered by the first radar. Cross-frame association matching is performed on the first object detected in each adjacent frame, and the number of successful matches is counted. When the number of successful cross-frame matches is greater than a preset threshold, the first object is determined to be a stable and valid target. Based on the temporal position points of the first object in multiple frames, the current motion trajectory of the first object in the first area is fitted and generated. By using the constraints of continuous cross-frame matching and the threshold of cross-frame matching count, false targets caused by false alarms, clutter, and instantaneous false detections of the first radar are suppressed, and real targets that are stable in time are selected to ensure the authenticity and validity of the generated first trajectory.

[0035] In some embodiments, the scheme of this embodiment can be combined with various optional schemes in one or more of the above embodiments. Before generating the first trajectory of the first object in the first region, the following steps are also included: receiving radar data output by a first radar, wherein the first radar is a millimeter-wave radar, the first radar periodically scans the first region, and the radar data is received by using a message queue and asynchronous worker thread parallel processing.

[0036] The first radar can be a millimeter-wave radar deployed for motion sensing detection of a first area. The first radar can continuously and cyclically scan and detect the first area at fixed time periods, ensuring continuous and uniform output of time-series data, achieving uninterrupted tracking, avoiding data gaps, trajectory jumps, and target loss, and ensuring the stability of subsequent trajectory generation. A message queue is used to queue and buffer the multi-frame radar data stream sent in real time by the first radar. Simultaneously, asynchronous worker threads are used in the background to perform data reception, unpacking, verification, and preprocessing operations in parallel, thereby obtaining a continuous and complete radar data stream. High-throughput access is achieved through the use of message queues and asynchronous worker threads for parallel processing, thereby reducing the risk of blocking and balancing real-time performance and stability.

[0037] S220. Associate the second trajectory of the first object in the first region with the first trajectory to obtain first information. The second trajectory is the movement trajectory of the first object in the first region before the first trajectory. The first trajectory and the second trajectory are the movement trajectories of the same first object. The first information indicates the trajectory, spatial position and speed of the first object.

[0038] The second trajectory can be a historical movement trajectory fragment of the same first object that has been generated and cached before the time node corresponding to the first trajectory. The second trajectory can save the historical movement trajectories of the same first pair before the first trajectory, which are used to splice and associate with the first trajectory in the current time period to form a long-term complete and continuous trajectory. Associating the second trajectory of the first object with the first trajectory in the first region includes: detecting whether the first object of the second trajectory and the first object of the first trajectory are the same first object, so that the two trajectories can be spliced ​​together temporally and spatially, solving the trajectory breakage problem caused by short-term occlusion or temporary disappearance of the first object, and achieving seamless continuation of the trajectory of the same first object.

[0039] In some embodiments, the scheme of this embodiment can be combined with various optional schemes in one or more of the above embodiments to generate a first trajectory of a first object in a first region, including the following steps: receiving multiple frames of adjacent radar data continuously output by a first radar in a first region; performing clustering and target detection on each frame of radar data to calculate the spatial position of the first object in each frame of radar data; performing cross-frame matching on the first object in the radar data corresponding to adjacent frames, and recording the number of successful cross-frame matchings for the same first object; if the number of successful cross-frame matchings is greater than a preset threshold, it is determined to be a stable and valid first object; fitting and smoothing the position and time sequence of the multiple frames of adjacent radar data of the same first object to generate a first trajectory.

[0040] In some embodiments, the scheme of this embodiment can be combined with various optional schemes in one or more of the above embodiments to generate a first trajectory of the first object in a first region, including the following steps: Based on the Euclidean nearest neighbor algorithm, cross-frame matching is performed on the first object in the radar data corresponding to adjacent frames continuously output by the first radar; based on the exponential moving average algorithm, the position of the first object that is successfully matched across frames is smoothed; in response to the existence of multiple first radar data in the radar data corresponding to adjacent frames continuously output by the first radar, the first trajectory formed by the position of the first object in the multiple first radar data is output.

[0041] Based on the Euclidean nearest neighbor algorithm, cross-frame matching is performed on the first object in the radar data corresponding to adjacent frames continuously output by the first radar. This includes: obtaining the radar data corresponding to the temporally adjacent frames continuously output by the first radar, detecting the first object contained in the radar data corresponding to each frame and the spatial location of the first object; using the Euclidean nearest neighbor algorithm, calculating the spatial Euclidean distance between the first objects contained in the radar data corresponding to adjacent frames; and establishing the matching relationship between the first objects contained in the radar data corresponding to adjacent frames based on the minimum distance as the matching criterion, thus completing the cross-frame matching association.

[0042] Based on the exponential moving average algorithm, the position of the first object successfully matched across frames is smoothed, including: retaining the spatial position of the same first object that is successfully matched across frames; and introducing exponential moving average filtering coefficients to smooth the spatial position of the same first object. When the first radar continuously outputs multiple frames of adjacent radar data with consecutive time sequences and there is a stable detectable first object, the frame-by-frame temporal position points of the first object after cross-frame matching and exponential moving average smoothing are statistically analyzed; the position coordinates are arranged sequentially according to time sequence, and the motion path of the first object is continuously fitted to generate and output the first trajectory.

[0043] For the first object that fails to match across multiple frames in a row, it can be eliminated based on the number of failed cross-frame matches or the trajectory survival time. Only when the number of successful cross-frame matches is greater than a preset threshold, a stable and valid trajectory of the first object is output. This can effectively suppress trajectory jitter and instantaneous position jumps caused by millimeter-wave radar detection, and improve the accuracy of subsequent role recognition and behavior feature recognition.

[0044] Associating the second trajectory of the first object with the first trajectory in the first region includes: estimating the motion speed and direction of the first object based on the position changes at previous and subsequent times, based on the first and second trajectories; performing spatiotemporal threshold constraint matching on the first and second trajectories, and integrating multiple dimensions such as actual position distance, predicted position distance, consistency of motion direction, and continuity of trajectory identifiers for comprehensive scoring and matching to generate a continuous and unique identifier; for the first object that is temporarily occluded or disappears, maintaining the trajectory loss state and supporting subsequent re-association and reconnection, effectively improving the continuity of identifiers and the integrity of the trajectory, and avoiding identity jumps and trajectory breaks.

[0045] S230. Generate second information corresponding to multiple first objects. The second information indicates the role recognition result of the first object. The second information is related to the spatiotemporal behavior characteristics of the first object. The spatiotemporal behavior characteristics of the first object are derived from the first information of the first object.

[0046] S240. Identify the third information of the second object in the second region. The second object is the object role among multiple first objects used to provide nursing services to the person being cared for. The third information indicates the spatiotemporal micro-motion characteristics of the second object in the second region. The second region is a local area in the first region used to characterize the second object performing nursing actions around the person being cared for.

[0047] S250. Based on the third and second information, detect whether the nursing services in the first area are compliant.

[0048] The technical solution of this invention relies on cross-frame matching and verification of adjacent radar data in multiple frames. Only after the number of matching hits meets the requirements is the first trajectory generated, and the second trajectory is associated with the current first trajectory. This effectively suppresses trajectory breaks and target jumps caused by radar false detections, clutter interference, and short-term blockages, and outputs standardized first information containing complete trajectory, spatial position, and motion speed. Based on the first information, the spatiotemporal behavioral characteristics of each first object are extracted. The second information corresponding to the role recognition result is generated based on the spatiotemporal behavioral characteristics. This eliminates the reliance on manual identity labeling and can stably distinguish between nursing service recipients, those being cared for, and other irrelevant personnel. A second area representing the bedside care space is delineated within the first area. Second objects with nursing service roles are selected from the first objects. The spatiotemporal micro-movement characteristics of the second objects within the second area are extracted to form the third information. The second information representing identity attributes and the third information representing fine care movement patterns are integrated. The compliance of nursing services is jointly determined from the dual dimensions of role identity and spatiotemporal micro-movement behavior. This avoids misjudgments and omissions based on single distance or duration judgments. It can identify situations such as genuine care, false stays, and lack of care, and realize intelligent, automated, and highly reliable compliance supervision of nursing services in home and elderly care scenarios.

[0049] Figure 3 This is a flowchart illustrating another nursing service detection method provided by an embodiment of the present invention. The technical solution of this embodiment further optimizes the process of generating second information corresponding to multiple first objects in the aforementioned embodiments based on the technical solutions of the above embodiments. This embodiment can be combined with various optional solutions in one or more of the above embodiments. For example... Figure 3 As shown, the nursing service detection method of this embodiment may include the following process: S310. Generate first information corresponding to multiple first objects within the first region. The first information indicates the trajectory, spatial position, and speed of the first object.

[0050] S320. Based on the first information of the first object, feature extraction is performed to obtain the spatiotemporal behavior features of the first object.

[0051] S330. Based on the spatiotemporal behavioral characteristics of the first object and the fourth information, generate a second information corresponding to the first object. The second information indicates the role type of the first object, and the fourth information indicates different types of spatiotemporal behavioral characteristics associated with different role types. The different types of spatiotemporal behavioral characteristics indicated by the fourth information include activity characteristics related to the object role used to provide nursing services to the nursing recipient, and activity characteristics related to the role of the nursing recipient.

[0052] The second information indicates the role recognition result of the first object. The second information is related to the spatiotemporal behavioral characteristics of the first object, and the spatiotemporal behavioral characteristics of the first object are derived from the first information of the first object.

[0053] By statistically analyzing and calculating the trajectory, position, and speed of the first object from its initial information, quantitative indicators characterizing the first object's behavioral patterns are extracted, yielding the spatiotemporal behavioral characteristics of the first object. These characteristics can describe the inherent behavioral features of the first object within a first region, distinguishing it from nursing staff, patients, and unrelated personnel. For example, the work behavior patterns of nursing staff can be accurately described through their spatial activity range, frequency of bedside visits, movement speed range, amplitude of positional changes, and nursing service-related care actions, thus naturally differentiating them from patients who are bedridden and have low activity levels.

[0054] In some embodiments, the solution in this embodiment can be combined with various optional solutions in one or more of the above embodiments, see [link to relevant documentation]. Figure 4 When the first object is a role for providing nursing services to a recipient, the spatiotemporal behavioral characteristics of the first object include at least one of the following: the activity range of the first object is distributed in the first area and the frequency of traveling back and forth in the second area is higher than the first frequency; the movement speed of the first object is greater than the first speed and less than the second speed; the position change of the first object is greater than the first change; or the first object performs nursing service-related actions in the first area; wherein, the activity intensity is used to characterize the first object's movement amplitude, displacement change and frequency of action per unit time; the first object's performance of nursing service-related actions in the first area is related to the height change corresponding to the spatial position of the first object in the first area as indicated in the first information.

[0055] When the first object is the role of the person providing nursing services to the person being cared for, the first object belongs to the nursing staff performing care, supervision, companionship, and bedside care work within the first area. The first object's activity range is distributed within the first area, and the frequency of movement to and from the second area is higher than the first area. This indicates that the first object can not only move around extensively within the first area, but also frequently enters and exits the second area and / or moves back and forth to and from the second area, with a higher number of movements to and from the second area than the first area. This distinguishes the first object from the elderly who remain immobile for long periods and rarely move to the bedside.

[0056] The first subject's movement speed is greater than the first speed but less than the second speed, indicating that the first subject's walking speed is neither too fast nor too slow. It is neither the extremely slow movement of an elderly person nor a fast run. Maintaining a moderate and even working gait, the role is distinguished by the movement rhythm. People who move too fast are irrelevant, and people who move too slowly are disabled elderly people. The caregivers whose gait characteristics match the care movements related to nursing services are identified.

[0057] The first subject's positional change is greater than the first change, indicating that the first subject's spatial coordinate displacement is large, movement is frequent, and positional fluctuations are significant per unit time. This is similar to the high activity level and constant movement of the caregivers, thus creating a strong contrast with the relatively fixed position of the bedridden elderly. Actions related to nursing services may include at least one of the following: typical micro-movements performed by the first subject in the second area, such as bending over, leaning forward, turning to the side, providing postural assistance, and tidying the bedding.

[0058] Activity intensity can be a comprehensive quantitative indicator used to quantify the overall amplitude of movement, displacement, and frequency of actions of the primary subject per unit time. It can standardize the quantification of the primary subject's activity level, turning "moving more" and "moving less" into calculable values, and helping to quickly distinguish caregivers from low-activity elderly individuals. Activity intensity quantifies the amplitude of movement, displacement fluctuation, and frequency of actions of the primary subject per unit time. Changes in spatial position corresponding to height can indicate the vertical undulations of the primary subject sensed by radar, corresponding to the height fluctuations caused by actions such as bending over, squatting, and standing up.

[0059] When the first object is identified as a nursing staff member providing nursing services to the patient, the spatiotemporal behavioral characteristics of the first object must meet at least the following combination of conditions: the first object's overall activity range covers the first area, and it repeatedly moves back and forth in the second area representing the bedside care space, with the first object's frequency of movement being higher than the preset first frequency; at the same time, the first object's movement speed is stably between the first speed and the second speed, exhibiting a medium-speed gait that conforms to the walking rhythm of nursing work; the first object's positional displacement change per unit time is greater than the first change threshold, indicating high overall activity level; in addition, the first object continuously generates fine care actions such as bending over and leaning over at the bedside within the first area, which are related to nursing work.

[0060] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments. When the first object is the role of the object being cared for, the spatiotemporal behavior characteristics of the first object include at least one of the following: the activity intensity of the first object is lower than the first activity intensity, the proportion of time the first object's activity range is in the second area is greater than the first proportion and the frequency of traveling back and forth in the second area is lower than the first frequency, the movement speed of the first object is less than the first speed, or the position change of the first object is less than the preset change.

[0061] The activity intensity of the first subject was lower than that of the first subject, indicating that the subject's overall limb activity and displacement were extremely low. Overall movements were slight, with little movement and few significant postural changes, and the overall activity level did not reach the intensity required for nursing staff to perform care-related actions. The first subject spent a greater proportion of time in the second area than in the first, and the frequency of moving back and forth within the second area was lower than in the first, indicating that the first subject spent the vast majority of their time within the second area where their bed was located, representing a high percentage of their dwell time. Simultaneously, they almost never frequently moved back and forth around the second area, exhibiting a state of prolonged confinement to the bed area with minimal movement and movement.

[0062] The first object's movement speed is less than the first speed, indicating that even with slight positional shifts, the overall movement rate remains low, characterized by slow, small-amplitude movements, far below the walking speed of caregivers performing nursing-related actions. The first object's positional change is less than the preset change, indicating that over a longer observation period, the first object's spatial coordinate shift is very small, its overall position is relatively fixed, its drift range is minimal, and there is no large-scale positional migration, consistent with the characteristics of a person receiving care.

[0063] By adopting the above scheme, a unique behavioral profile of the person being cared for is constructed through multi-dimensional quantitative constraints such as activity level, spatial dwell characteristics, regional travel behavior, movement speed, and positional stability. As long as any one of the characteristic conditions is met, the person can be included in the category of the person being cared for. The combination of multiple conditions can further improve the accuracy of role recognition and avoid misjudging caregivers and family members as the person being cared for.

[0064] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments, and after generating the second information corresponding to multiple first objects, the solution further includes the following steps: The second information identified at different times is input into the role state machine. The role state machine is used to stabilize the role identification result indicated by the second information to suppress the instantaneous jitter and frequent switching of role identification. The role state machine includes at least one of the following: candidate state, locked state, soft state, lost state and expired state, which correspond to the role's pending confirmation state, stable locked state, fluctuation buffer state, temporary lost state and expired state, respectively.

[0065] The second information, generated independently frame by frame at each moment, is fed into a pre-defined character state machine for unified timing management and stability control, instead of directly using the single-frame character recognition result as the final identity output. The character state machine, through multiple internal working states combined with timing judgment rules, performs smoothing, anti-jitter, delayed confirmation, and delayed cancellation processing on the character recognition result, effectively suppressing instantaneous jitter, jumps, and frequent switching of the character caused by environmental noise, action changes, and occlusion interference. The character state machine can be configured with at least one type of working state, which can be a candidate state, a locked state, a soft state, a lost state, and an expired state, corresponding to the character's pending confirmation state, stable locked state, fluctuation buffer state, temporary lost state, and expired invalid state, respectively, thereby achieving orderly, controllable, and stable management of the entire lifecycle of the target character.

[0066] The candidate state indicates that the first object has been detected and its role type has been preliminarily determined, but has not yet reached a stable and reliable level, and is in a state of pending confirmation. The locked state indicates that after verification across multiple consecutive frames, the role recognition results are consistent and the confidence level meets the set conditions, the role type has been stably confirmed and locked, and will not be easily changed. The soft state indicates that the role recognition results have experienced short-term fluctuations or the confidence level is between reliable and unreliable; the role is not immediately switched, and the original determination is maintained to avoid jitter. The lost state indicates that the first object has temporarily left the detection area or is occluded and cannot be observed, but the role and trajectory information of the first object are still retained, waiting for the first object to reappear. The expired state indicates that the first object has been missing for a longer period than a preset threshold, and it is determined that the first object has left the monitoring range; the corresponding trajectory and role resources will be released.

[0067] By adopting the above scheme, instead of directly using the single-frame recognition result as the final role recognition result, the role recognition results of each person in a continuous time sequence are input into a preset role state machine. The state machine is used to stabilize and smooth the role judgment results, thereby improving the continuity and reliability of role recognition.

[0068] S340. Identify the third information of the second object in the second region. The second object is the object role among multiple first objects used to provide nursing services to the nursing recipient. The third information indicates the spatiotemporal micro-motion characteristics of the second object in the second region. The second region is a local area in the first region used to characterize the second object performing nursing actions around the nursing recipient.

[0069] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments. The setting of the second region is related to the fifth information. The fifth information includes the bed position of the person being cared for, the range of the bedside area, and the position where the person performs nursing service-related actions in the first region. The third information includes at least one of the following: the percentage of time the second person spends in the second region, the step length and step frequency of the second person in the second region, the directional entropy of the second person in the second region, or the drift amount of the second person in the second region. The directional entropy is used to measure the directional diversity of the second person performing nursing service-related actions in the first region, the offset is used to constrain the nursing service-related actions to occur in the second region, and the step length and step frequency are used to identify continuous nursing service-related actions.

[0070] The second area can be a localized care area surrounding the bed, defined within the first area. This second area filters out distractions from corridors, passageways, and long walks, retaining only the area where care-related actions are likely to occur. The percentage of time the second subject spends in the second area represents the proportion of their total activity time within the statistical period. This percentage can be used to determine whether caregivers are truly on duty and consistently present at the bedside, eliminating false reports of attendance such as passing by or making brief stops.

[0071] The second subject's stride length in the second area can be defined as the distance covered in a single movement, while the gait frequency can be defined as the number of steps taken per unit of time, both reflecting the subject's small-scale walking rhythm. By using the stride length and gait frequency functions in the second area, we can identify care-related actions such as pacing back and forth around the bedside, small movements, and repeated standing within the second area, distinguishing between normal large strides and genuine care micro-movements. Gait frequency and stride length are used to identify continuous, small-scale, high-frequency care-like movements.

[0072] The directional entropy of a second object in the second area can be an indicator of the richness of its movement direction variations. Caregiving actions related to nursing services are typically multi-directional, back-and-forth micro-movements; the more varied and numerous the directional movements, the higher the directional entropy. Conversely, unidirectional straight movement results in low directional entropy. The directional entropy of a second object in the second area can be used to prevent misjudging unidirectional walking or passing through a corridor as nursing actions. Directional entropy measures whether the movement directions are sufficiently diverse to prevent misjudging unidirectional movement as nursing actions. The drift amount of a second object in the second area can be the range of fluctuation in its overall position within the area, reflecting whether the activity is concentrated in a localized area. The drift amount is used to constrain actions to occur within a localized area, preventing entire stretches of walking from being misjudged as nursing actions.

[0073] S350. Based on the third and second information, detect whether the nursing services in the first area are compliant.

[0074] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments. Based on the third information and the second information, detecting whether the nursing service in the first area is compliant may include the following steps: generating sixth information, which is related to the third information and the second information. The sixth information indicates whether a nursing relationship is formed between the second object and the third object, and the third object is the nursing object among multiple first objects; detecting whether the nursing service in the first area is compliant based on the sixth information.

[0075] The second object can be a person whose role identification result among multiple first objects is a nursing service provider, i.e., a caregiver. The third object can be a care recipient whose role identification result among multiple first objects is a person who needs to receive care-related actions related to nursing services, typically a bedridden or disabled elderly person. As a recipient of nursing services, this person is used to match with a caregiver and form a corresponding care relationship. The sixth information can be a judgment result generated based on the role identification record of the second information and the spatiotemporal micro-movements indicated by the third information, the distance between the second and third objects, and the positional relationship between the second and third objects. This result is used to indicate whether an effective care relationship has been established between the caregiver and the care recipient.

[0076] Based on the role recognition results in the second information, a second object (as a caregiver) and a third object (as a patient) are distinguished. Simultaneously, combining the spatiotemporal micro-movement characteristics of the second object within the second area from the third information, it is determined whether the caregiver is close to the bed area, whether they remain there for a sufficient duration, whether they exhibit small-scale, high-frequency nursing-related micro-movements, and whether they are in close proximity to the patient. The sixth information is generated by comprehensively considering identity matching conditions, spatial distance conditions, regional location conditions, and spatiotemporal micro-movement authenticity. This sixth information is used to definitively determine whether a genuine and valid one-to-one nursing pairing relationship has been established between the second and third objects.

[0077] Based on the nursing relationship determination result of the sixth piece of information, the compliance of the overall nursing service behavior in the first area is determined. If the sixth piece of information indicates that the second and third objects have formed a stable and effective nursing relationship, and the second object continuously exhibits compliant nursing micro-actions in the second area, then the current nursing service is determined to be genuine and effective, and the nursing service is compliant. If the sixth piece of information shows that the second and third objects have not formed an effective nursing relationship, or although the second object is present, there are no care actions related to the nursing service, and the object only makes a false stop, then the nursing service is determined to be inadequate and non-compliant.

[0078] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments. The sixth information is also related to the seventh information. The seventh information indicates the distance between the second object and the third object, the duration of the micro-action related to the spatiotemporal micro-action characteristics indicated by the third information, the intensity of the micro-action related to the spatiotemporal micro-action characteristics indicated by the third information, and the interaction state between the second object and the third object in the second region.

[0079] When generating the sixth piece of information for determining nursing relationships, it not only combines the role type of the second piece of information and the spatiotemporal micro-motion characteristics of the third piece of information, but can also further integrate the multi-dimensional interaction parameters of the seventh piece of information. This involves comprehensive verification from four aspects: the spatial distance between the second and third objects, the duration and intensity of the spatiotemporal micro-motions related to nursing services, and the regional interaction status. This improves the objectivity and accuracy of nursing relationship judgment and avoids misjudgments such as false nursing care or accidental proximity. The distance between the second and third objects can be the straight-line distance between them at the same moment, used to determine whether the second and third objects are within an effective close-range range where care actions related to nursing services can be carried out. If the distance is too far, an effective nursing relationship cannot be formed.

[0080] The duration of micro-movements related to spatiotemporal micro-movement characteristics can be the cumulative duration of continuous care-related actions performed by the second subject within the second area. This duration can be used to exclude short-term disruptive behaviors such as momentary passing or brief pauses. The intensity of micro-movements related to spatiotemporal micro-movement characteristics can be a quantification of the amplitude and activity level of micro-movements such as small-range shifting, back-and-forth adjustments, and bending over within the second area. This is used to distinguish genuine care-related actions from inactive, fictitious on-duty behaviors such as remaining still, daydreaming, or simply standing. The interaction state between the second and third subjects within the second area can be defined as a coexistence state where the second and third subjects are simultaneously located within the second area, their positions corresponding and their spatial relationship stable. This ensures that care-related actions occur within the second area.

[0081] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments. Detecting whether the nursing services in the first area are compliant based on the sixth information may include the following steps: In response to the sixth information indication that a nursing relationship is formed, a first nursing event is generated based on the nursing relationship between the second and third objects based on the sixth information indication; in response to the sixth information indication that no nursing relationship is formed, a second nursing event is generated based on the second object's continuous nursing service-related actions in the second area; the compliance of nursing services in the first area is detected based on the first or second nursing event.

[0082] When the sixth information determines that the distance between the second and third objects is compliant, the positions match, the roles are stable, and the interactions are normal, thus successfully forming a reliable nursing pairing relationship, a first nursing event is directly generated. The first nursing event represents a presence of people, a pairing, and a scene, constituting a complete, reliable, and authentic standard care behavior record. If, due to temporary obstruction, unstable proximity of personnel, or short-term role fluctuations, a pairing between caregiver and recipient fails, and the sixth information determines that no nursing relationship has been formed, it is not directly judged as lacking care. Instead, it continues to detect whether the second object continuously exhibits small-scale, high-frequency, multi-directional micro-movements within the second area that conform to nursing characteristics. If the movements are authentic and continuously compliant, a second nursing event is generated as a fallback, retained as weak evidence to avoid missed judgments due to pairing failures. In this way, a comprehensive judgment can be made regarding whether a valid nursing event exists at the current moment: as long as a highly credible first nursing event exists or a fallback second nursing event exists, it can be determined that a genuine care behavior has occurred on-site, and the nursing service is deemed compliant; if neither type of event exists, it is determined that there is no effective care and the nursing service is non-compliant.

[0083] For example, see Figure 5 When a stable elderly-caregiver pairing cannot be formed, or when sufficient hard-pair evidence is lacking, if a continuous, small-range, high-frequency care-like movement is detected within the core care area, the core care activity event will still be output as a weak evidence event. This event will be supplemented with evidence source markers, anchor point locations, mean values ​​of movement features, and confidence information. The event_engine_v2_care already has an explicit fallback event engine, which can still identify core_zone_care_activity even when pairing is incomplete. Furthermore, it reduces false alarms through stricter minimum duration thresholds and post-processing priority control.

[0084] The above scheme achieves a dual-path fusion judgment of strong and weak evidence, relying on both the pairing of the second and third parties and real actions, which greatly improves the completeness of care event identification. It effectively solves the problem of nursing relationship pairing failure caused by short-term occlusion by personnel, temporary loss of the target, and role shaking, and avoids the situation of missed detection where real care is judged as no care. Strong evidence is used for normal pairing, and actions are used as a fallback for abnormal pairing. It improves the environmental adaptability in complex indoor care scenarios, makes compliance judgments reasonable and verifiable, and makes the results traceable and reviewable, thus meeting the requirements of intelligent supervision of long-term care.

[0085] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments to aggregate adjacent or overlapping first and second nursing events into a nursing session, and record: session start and end time; session duration; whether there is evidence of a first nursing event that forms a nursing relationship between the second object and the third object; whether there is evidence of care actions related to nursing services in the first area; whether there is evidence of a second nursing event that does not form a nursing relationship between the second object and the third object, and the source of the evidence. Then, checks are performed based on preset long-term care insurance compliance rules, such as: whether the second object is in the second area; whether the third object is in the second area; whether the second object and the third object have formed a stable nursing relationship; whether the second object has continuous care actions related to nursing services in the first area; whether the second object has reached the preset minimum nursing duration; whether only the second nursing event exists between the second object and the third object without the first nursing event; and whether there is obvious abnormal interruption or abnormally short session. Finally, the conclusions such as "compliant", "suspected non-compliant", or "requires manual review" are output.

[0086] In some embodiments, the solution in this embodiment can be combined with various optional solutions in one or more of the above embodiments to output nursing events, nursing sessions and compliance judgment results as structured documents, timeline reports or audit logs; in the event of a dispute, packet capture playback, database history rerun and timeline visualization can be used for offline review.

[0087] The technical solution of this invention relies on radar perception within a first area to generate first information consisting of the trajectory, spatial position, and speed of a first object. Reliable spatiotemporal behavioral features are extracted, and combined with fourth information—standard spatiotemporal behavioral features of multiple roles—to complete role recognition of the first object within the first area, distinguishing between nursing service roles and the roles of those being cared for, thus avoiding interference from manual calibration and single-frame recognition jitter. A second area is specifically defined, focusing on extracting refined spatiotemporal micro-motion features from nursing staff to generate third information. Through spatial area constraints, irrelevant interference behaviors such as passage through corridors, long-distance loitering, and aimless wandering are effectively filtered out, eliminating invalid non-nursing activity data and ensuring the targetedness and effectiveness of action feature extraction. Based on this, the second information of role recognition and the third information of the second object's spatiotemporal micro-motions in the second area are integrated, replacing the traditional single-distance judgment method. This effectively avoids problems such as false on-duty status, invalid loitering, and misjudgment of passing by, significantly improving the accuracy, stability, and intelligent supervision efficiency of nursing service compliance detection in long-term care scenarios, and adapting to complex indoor home care application scenarios.

[0088] In some embodiments, Figure 6This is a schematic diagram of a nursing service testing device provided in an embodiment of the present invention. This embodiment is applicable to situations where compliance checks are performed on nursing services, especially in cases where compliance checks are performed on nursing services during long-term care or long-term nursing. The nursing service testing device can be implemented in hardware and / or software and can be configured in an electronic device. Figure 6 As shown, the nursing service detection method of this embodiment may include the following: The generation module 610 is used to generate first information corresponding to multiple first objects within a first region, wherein the first information indicates the trajectory, spatial position, and velocity of the first object. The generation module 610 is also used to generate second information corresponding to the plurality of first objects, the second information indicating the role recognition result of the first object, the second information being related to the spatiotemporal behavior characteristics of the first object, and the spatiotemporal behavior characteristics of the first object originating from the first information of the first object; The identification module 620 is used to identify third information of a second object in a second region. The second object is an object role among the plurality of first objects that provides nursing services to the object being cared for. The third information indicates the spatiotemporal micro-motion characteristics generated by the second object in the second region. The second region is a local region in the first region that represents the second object performing nursing actions around the object being cared for. The detection module 630 is used to detect whether the nursing services in the first area are compliant based on the third information and the second information.

[0089] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments to generate first information corresponding to multiple first objects in the first region, including: Generate a first trajectory of the first object in a first region. The first trajectory is related to the trajectory formed by the position of the first object in multiple first radar data. The multiple first radar data are radar data corresponding to multiple adjacent frames continuously output by the first radar in the first region. The first object in each adjacent frame of the multiple first radar data is successfully matched across frames and the number of successful cross-frame matching is greater than a preset number threshold. The second trajectory of the first object in the first region is associated with the first trajectory to obtain the first information. The second trajectory is the movement trajectory of the first object in the first region before the first trajectory. The first trajectory and the second trajectory are the movement trajectories of the same first object.

[0090] In some embodiments, the scheme of this embodiment can be combined with various optional schemes in one or more of the above embodiments, and before generating the first trajectory of the first object in the first region, it further includes: The radar data output by the first radar, which is a millimeter-wave radar, is received. The first radar periodically scans the first area. The radar data is received by using a message queue and asynchronous worker threads for parallel processing.

[0091] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments, and generating the first trajectory of the first object in the first region includes: Based on the Euclidean nearest neighbor algorithm, cross-frame matching is performed on the first object in the radar data corresponding to adjacent frames continuously output by the first radar. Based on the exponential moving average algorithm, the position of the first object that is successfully matched across frames is smoothed. In response to the existence of multiple first radar data in adjacent frames of radar data continuously output by the first radar, the first trajectory formed by the position of the first object in the multiple first radar data is output.

[0092] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments, and the generation of the second information corresponding to the plurality of first objects includes: Based on the first information of the first object, feature extraction is performed to obtain the spatiotemporal behavioral features of the first object; Based on the spatiotemporal behavioral characteristics of the first object and the fourth information, second information corresponding to the plurality of first objects is generated. The second information indicates the role type of the first object, and the fourth information indicates different types of spatiotemporal behavioral characteristics associated with different role types. The different types of spatiotemporal behavioral characteristics indicated by the fourth information include activity characteristics related to the object role used to provide nursing services to the nursing recipient, and activity characteristics related to the role of the nursing recipient.

[0093] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments. When the first object is an object role for providing nursing services to the object being cared for, the spatiotemporal behavioral characteristics of the first object include at least one of the following: the activity range of the first object is distributed in the first area and the frequency of traveling back and forth in the second area is higher than the first frequency; the movement speed of the first object is greater than the first speed and less than the second speed; the position change of the first object is greater than the first change; or the first object performs nursing service-related actions in the first area. The activity intensity is used to characterize the amplitude of movement, displacement change, and frequency of actions of the first object per unit time. The nursing service-related actions performed by the first object in the first area are related to the height change corresponding to the spatial position of the first object in the first area as indicated in the first information. When the first object is a person being cared for, the spatiotemporal behavioral characteristics of the first object include at least one of the following: the activity intensity of the first object is lower than the first activity intensity; the proportion of time the first object spends in the second area is greater than the first proportion and the frequency of traveling back and forth in the second area is lower than the first frequency; the movement speed of the first object is less than the first speed; or the position change of the first object is less than the preset change.

[0094] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments, and after generating the second information corresponding to the plurality of first objects, it further includes: The second information identified at different times is input into the character state machine, which is used to stabilize the character recognition result indicated by the second information to suppress instantaneous jitter and frequent switching of character recognition. The role state machine includes at least one of the following: candidate state, locked state, soft state, lost state, and expired state, which correspond to the role's pending confirmation state, stable locked state, fluctuation buffer state, temporary lost state, and expired state, respectively.

[0095] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments. The setting of the second region is related to the fifth information, which includes the bed position of the cared-for object, the range of the bedside area, and the position where the nursing service-related actions are performed in the first region. The third information includes at least one of the following: the percentage of time the second object spends in the second region, the step length and step frequency of the second object in the second region, the directional entropy of the second object in the second region, or the drift amount of the second object in the second region. The directional entropy is used to measure the directional diversity of the second object performing nursing service-related actions in the first region, the drift amount is used to constrain nursing service-related actions to occur in the second region, and the step length and step frequency are used to identify continuous nursing service-related actions.

[0096] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments to detect whether the nursing services in the first area are compliant based on the third information and the second information, including: A sixth piece of information is generated, which is related to the third information and the second information. The sixth piece of information indicates whether a nursing relationship is formed between the second object and the third object, and the third object is the object being cared for among the plurality of first objects. Based on the sixth piece of information, the compliance of the nursing services in the first area is determined.

[0097] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments. The sixth information is also related to the seventh information. The seventh information indicates the distance between the second object and the third object, the duration of the micro-action related to the spatiotemporal micro-action characteristics indicated by the third information, the intensity of the micro-action related to the spatiotemporal micro-action characteristics indicated by the third information, and the interaction state of the second object and the third object in the second region.

[0098] In some embodiments, the solution of this embodiment can be combined with various optional solutions in one or more of the above embodiments to detect whether the nursing services in the first area are compliant based on the sixth information, including: In response to a sixth information instruction to form a nursing relationship, a first nursing event is generated based on the nursing relationship between the second object and the third object according to the sixth instruction; In response to the sixth information indicating that no nursing relationship has been formed, a second nursing event is formed based on the second object's continuous nursing service-related actions in the second area; The compliance of nursing services in the first area is determined based on the first nursing event or the second nursing event.

[0099] The technical solution of this invention generates first information about multiple first objects within a first region. This first information includes the trajectory, spatial location, and velocity of the first objects, enabling a quantitative representation of the motion state of the first objects within the first region. Based on this first information, spatiotemporal behavioral features of the first objects are extracted. Then, based on these spatiotemporal behavioral features, a role recognition result strongly correlated with the behavioral patterns of the first objects is generated. Quantifiable spatiotemporal behavioral features serve as the basis for role classification, accurately distinguishing the roles of each first object within the first region and providing a basis for screening nursing service-related objects. A second region representing the care space is delineated from the first region, and second objects belonging to nursing service roles are identified. These second objects are specifically extracted for providing nursing services to the recipients. The second object in the care service exhibits spatiotemporal micro-motion characteristics within the second area, enabling layered perception from macroscopic trajectories to refined micro-movements within the care area. This effectively filters out irrelevant walking behaviors and accurately captures the characteristics of real nursing operations, overcoming the limitation of relying solely on macroscopic movements to identify refined nursing actions. Furthermore, by integrating role recognition results with the spatiotemporal micro-motion characteristics of the second object in the second area for joint judgment, the system confirms that the object possesses the identity attributes of a nursing staff member and verifies the existence of micro-movement behaviors that conform to nursing characteristics within the care area. This significantly reduces the probability of misjudgments and omissions in nursing services, enabling objective, stable, and automated completion of nursing service compliance detection within the first area. This effectively improves the completeness, accuracy, and reliability of nursing service supervision in care scenarios.

[0100] The nursing service testing device provided in this embodiment of the invention can execute the nursing service testing method provided in any of the above embodiments of the invention, and has the corresponding functions and beneficial effects of executing the nursing service testing method. For details, please refer to the relevant operations of the nursing service testing method in the foregoing embodiments.

[0101] It is worth noting that the various units and modules included in the above-mentioned device are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the protection scope of the embodiments of the present invention.

[0102] In one embodiment, Figure 7 This is a structural block diagram of an electronic device provided in an embodiment of the present invention, such as... Figure 7The diagram illustrates a schematic representation of an electronic device 10 that can be used to implement embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0103] like Figure 7 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor 11 to enable the at least one processor 11 to perform the nursing service detection method provided by the present invention.

[0104] The processor 11 can perform various appropriate actions and processes based on a computer program stored in the read-only memory (ROM) 12 or a computer program loaded from the storage unit 18 into the random access memory (RAM) 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0105] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0106] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the nursing service detection method provided by this invention.

[0107] In some embodiments, the nursing service detection method provided herein may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the nursing service detection method by any other suitable means (e.g., by means of firmware).

[0108] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard parts (ASSPs), systems-on-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0109] Computer programs used to implement the nursing service detection method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0110] In the context of this invention, a computer-readable storage medium stores computer instructions that are used to cause a processor to execute and implement the nursing service detection method provided by this invention.

[0111] The present invention also provides a computer program product comprising a computer program that, when executed by a processor, implements the nursing service detection method provided according to embodiments of the present invention. A computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. The computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0112] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0113] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0114] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0115] This invention also provides a computer program product, including a computer program that, when executed by a processor, can implement the nursing service detection method provided in any embodiment of this application.

[0116] In the implementation of the computer program product, computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone 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 remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0117] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0118] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting nursing services, characterized in that, The method includes: Generate first information corresponding to multiple first objects within a first region, wherein the first information indicates the trajectory, spatial position, and velocity of the first object; Generate second information corresponding to the plurality of first objects. The second information indicates the role recognition result of the first object. The second information is related to the spatiotemporal behavioral characteristics of the first object. The spatiotemporal behavioral characteristics of the first object are derived from the first information of the first object. The third information of the second object in the second region is identified. The second object is the object role among the plurality of first objects used to provide nursing services to the object being cared for. The third information indicates the spatiotemporal micro-motion characteristics of the second object in the second region. The second region is a local area in the first region used to characterize the second object performing nursing actions around the object being cared for. Based on the third information and the second information, the compliance of the nursing services in the first area is detected.

2. The method according to claim 1, characterized in that, Generate first information corresponding to multiple first objects within the first region, including: Generate a first trajectory of the first object in a first region. The first trajectory is related to the trajectory formed by the position of the first object in multiple first radar data. The multiple first radar data are radar data corresponding to multiple adjacent frames continuously output by the first radar in the first region. The first object in each adjacent frame of the multiple first radar data is successfully matched across frames and the number of successful cross-frame matching is greater than a preset number threshold. The second trajectory of the first object in the first region is associated with the first trajectory to obtain the first information. The second trajectory is the movement trajectory of the first object in the first region before the first trajectory. The first trajectory and the second trajectory are the movement trajectories of the same first object.

3. The method according to claim 2, characterized in that, Before generating the first trajectory of the first object in the first region, the method further includes: The radar data output by the first radar, which is a millimeter-wave radar, is received. The first radar periodically scans the first area. The radar data is received by using a message queue and asynchronous worker threads for parallel processing.

4. The method according to claim 2, characterized in that, Generating the first trajectory of the first object in the first region includes: Based on the Euclidean nearest neighbor algorithm, cross-frame matching is performed on the first object in the radar data corresponding to adjacent frames continuously output by the first radar. Based on the exponential moving average algorithm, the position of the first object that is successfully matched across frames is smoothed. In response to the existence of multiple first radar data in adjacent frames of radar data continuously output by the first radar, the first trajectory formed by the position of the first object in the multiple first radar data is output.

5. The method according to claim 1, characterized in that, The generation of the second information corresponding to the plurality of first objects includes: Based on the first information of the first object, feature extraction is performed to obtain the spatiotemporal behavioral features of the first object; Based on the spatiotemporal behavioral characteristics of the first object and the fourth information, second information corresponding to the plurality of first objects is generated. The second information indicates the role type of the first object, and the fourth information indicates different types of spatiotemporal behavioral characteristics associated with different role types. The different types of spatiotemporal behavioral characteristics indicated by the fourth information include activity characteristics related to the object role used to provide nursing services to the nursing recipient, and activity characteristics related to the role of the nursing recipient.

6. The method according to claim 5, characterized in that, When the first object is a role for providing nursing services to a recipient, the spatiotemporal behavioral characteristics of the first object include at least one of the following: the activity range of the first object is distributed in the first area and the frequency of traveling back and forth in the second area is higher than the first frequency; the movement speed of the first object is greater than the first speed and less than the second speed; the position change of the first object is greater than the first change; or the first object performs nursing service-related actions in the first area; wherein, the activity intensity is used to characterize the range of motion, displacement change, and frequency of actions of the first object per unit time; the nursing service-related actions performed by the first object in the first area are related to the height change corresponding to the spatial position of the first object in the first area as indicated in the first information; When the first object is a person being cared for, the spatiotemporal behavioral characteristics of the first object include at least one of the following: the activity intensity of the first object is lower than the first activity intensity, the proportion of time the first object spends in the second area is greater than the first proportion and the frequency of traveling back and forth in the second area is lower than the first frequency, the movement speed of the first object is less than the first speed, or the position change of the first object is less than the preset change.

7. The method according to claim 5, characterized in that, After generating the second information corresponding to the plurality of first objects, the method further includes: The second information identified at different times is input into the character state machine, which is used to stabilize the character recognition result indicated by the second information to suppress instantaneous jitter and frequent switching of character recognition. The role state machine includes at least one of the following: candidate state, locked state, soft state, lost state, and expired state, which correspond to the role's pending confirmation state, stable locked state, fluctuation buffer state, temporary lost state, and expired state, respectively.

8. The method according to claim 1, characterized in that, The setting of the second area is related to the fifth information, which includes the bed position of the person being cared for, the range of the bedside area, and the position where the action related to nursing services is performed in the first area; the third information includes at least one of the following: the percentage of time the second person spends in the second area, the step length and step frequency of the second person in the second area, the directional entropy of the second person in the second area, or the amount of drift of the second person in the second area; The directional entropy is used to measure the directional diversity of the second object performing nursing service-related actions in the first region, the offset is used to constrain nursing service-related actions to occur in the second region, and the step length and step frequency are used to identify continuous nursing service-related actions.

9. The method according to claim 1, characterized in that, Based on the third information and the second information, the compliance of the nursing services in the first area is detected, including: A sixth piece of information is generated, which is related to the third information and the second information. The sixth piece of information indicates whether a nursing relationship is formed between the second object and the third object, and the third object is the object being cared for among the plurality of first objects. Based on the sixth piece of information, the compliance of the nursing services in the first area is determined.

10. The method according to claim 9, characterized in that, The sixth information is also related to the seventh information, which indicates the distance between the second object and the third object, the duration of the micro-action related to the spatiotemporal micro-action characteristics indicated by the third information, the intensity of the micro-action related to the spatiotemporal micro-action characteristics indicated by the third information, and the interaction state between the second object and the third object in the second region.

11. The method according to claim 9, characterized in that, Detecting whether the nursing services in the first area are compliant based on the sixth piece of information includes: In response to a sixth information instruction to form a nursing relationship, a first nursing event is generated based on the nursing relationship between the second object and the third object according to the sixth instruction; In response to the sixth information indicating that no nursing relationship has been formed, a second nursing event is formed based on the second object's continuous nursing service-related actions in the second area; The compliance of nursing services in the first area is determined based on the first nursing event or the second nursing event.

12. A nursing service detection device, characterized in that, The device includes: The generation module is used to generate first information corresponding to multiple first objects within a first region, wherein the first information indicates the trajectory, spatial position, and velocity of the first object. The generation module is also used to generate second information corresponding to the plurality of first objects. The second information indicates the role recognition result of the first object. The second information is related to the spatiotemporal behavior characteristics of the first object. The spatiotemporal behavior characteristics of the first object are derived from the first information of the first object. The identification module is used to identify third information of a second object in a second region. The second object is an object role among the plurality of first objects that provides nursing services to the object being cared for. The third information indicates the spatiotemporal micro-motion characteristics of the second object in the second region. The second region is a local area in the first region that represents the second object performing nursing actions around the object being cared for. The detection module is used to detect whether the nursing services in the first area are compliant based on the third information and the second information.

13. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1-11.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to perform the method of any one of claims 1-11.

15. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method according to any one of claims 1-11.

Images